liens code de la propriété intellectuelle. articles l 122....
TRANSCRIPT
AVERTISSEMENT
Ce document est le fruit d'un long travail approuvé par le jury de soutenance et mis à disposition de l'ensemble de la communauté universitaire élargie. Il est soumis à la propriété intellectuelle de l'auteur. Ceci implique une obligation de citation et de référencement lors de l’utilisation de ce document. D'autre part, toute contrefaçon, plagiat, reproduction illicite encourt une poursuite pénale. Contact : [email protected]
LIENS Code de la Propriété Intellectuelle. articles L 122. 4 Code de la Propriété Intellectuelle. articles L 335.2- L 335.10 http://www.cfcopies.com/V2/leg/leg_droi.php http://www.culture.gouv.fr/culture/infos-pratiques/droits/protection.htm
U. F. R. ENSTIB Ecole Doctorale Sciences et Ingénierie des Ressources Procédés Produits et Environnement Département de Formation Doctorale Sciences du Bois
Thèse
présentée pour l'obtention du titre de
Docteur de l'Université Henri Poincaré, Nancy-1
Mention Sciences du Bois
par
Francis MBURU
Etude et Valorisation de différents bois du Kenya -----------
Study and valorization of different Kenyan wood species
Soutenue publiquement le 12 juin 2007 devant la commission d'examen :
Membres du jury : Rapporteurs : M. A. CASTELLAN Professeur, Université de Bordeaux I M. J.H. THOMASSIN Professeur, Université de Poitiers Président : M. X. DEGLISE Professeur émérite, Université Henri Poincaré Nancy 1 Examinateurs : Mme T. STEVANOVIC Professeur, Université Laval de Québec M. P. GERARDIN Université Henri Poincaré Nancy 1 -------------------------------------------------------------------------------------------------------------------------------
Laboratoire d’Etudes et de Recherches sur le Matériau Bois UMR 1093 INRA/ENGREF/UHP Faculté des Sciences et Techniques, 54506 Vandoeuvre les Nancy, France
Acknowledgements
I wish to express my sincere gratitude to Professor Philippe GERARDIN, in charge of
Organic Chemistry and Microbiology team from the LERMAB, for accepting to supervise me
for Ph.D studies. I thank him for his personal involvement and interest in the research work.
Many times, he went out of his way to solve problems that occurred during my three years of
study in Nancy and Moi University. He read and corrected my dissertation to its present form.
Many thanks go to Dr. Mathieu PETRISSANS for his guidance during heat treatment using
the oven reactor. His assistance extended to preparation of samples and testing the contact
angle of heat treated wood using goniometer.
Thanks to Dr. Stéphane DUMARCAY for his assistance in the chemical analysis of wood
extracts and heat treated wood using NMR spectroscopy.
I would wish to thank Professor A. ZOULALIAN and Professor A. MERLIN for
administrative organization within the laboratories.
I wish to thank Ms C. ANTONI, the secretary in our laboratory for coordinating with
Professor and other relevant authorities for the benefit of all doctoral students in the
laboratory. Thanks to Mr Ludovic MOUTON for his assistance in preparation of laboratory
equipments.
I am truly indebted to Professor Xavier DEGLISE to whom I wrote my first mail enquiring
possibilities of studying in Nancy University in the year, 2002.
Thanks to all the staff of ENSTIB, especially Doctor J.F. BOCQUET, who assisted me to
perform mechanical tests of heat treated wood. Also, I cannot forget to thank Ms Françoise
HUBER of ENGREF for assisting me to perform electronic microscopy.
I thank my family for the trust they bestowed on me during the time of studies in France.
I wish to sincerely thank the French government for funding my studies. This made it possible
to realize my dream of obtaining PhD, which is very important for career development and
promotion at Moi University.
Last but not the least, I wish to thank all those who assisted me in one way or another during
my studies here in France and in Kenya. To all my colleagues in the laboratory, Ambrose
KIPROP, Thierry KOUMBI, Bouddah POATY, Thierry LEKOUNOUGOU, Steeve
MOUNGUENGUI, Gildas NGUILA, Peter SIRMA, Kamal IAYCH, Mounir CHAOUCH and
Joseph-Privat ONDO. I thank you all.
Table of content
Introduction générale 1
General introduction 4
Part 1. Effect of extractives on natural durability of Prunus africana
wood species 7
1. Context of study 7
1.1. Natural durability of wood 7
1.2. What is known about Prunus africana 8
2. Literature review 13
2.1. Wood biodegradation 13
2.2. Chemical nature of extractives associated with wood durability 14
2.3. Wood attacking microorganisms 18
2.3.1. Moulds 19
2.3.2. Sapstain 20
2.3.3. Brown rot 20
2.3.4. White rot 21
2.3.5. Soft rot 21
2.3.6. Bacteria 22
2.4. Termites 23
2.5. Other wood degraders 24
2.5.1. Insects 24
2.5.2. Marine borers 24
3. Materials and methods 26
3.1. Wood species utilized 26
3.2. Measurement of wood extractives 26
3.2.1. Soxhlet extraction 26
3.2.2. Quantification of extractive 27
3.3. Biological tests 27
3.3.1. Prunus africana natural durability 27
3.3.2. Growth inhibition 29
3.3.3. Measurement of efficacy against termites 29
3.4. Determination of dimensional stability 31
3.5.1. Determination of Klason lignin content 31
3.5.2. Determination of mineral content 32
3.6. Scanning electron microscopy 32
3.7. Spectroscopic analysis 32
3.7.1. NMR Analysis 32
3.7.2. Infrared analysis 33
4. Results and discussion 34
4.1. Soxhlet extraction 34
4.2. Biological tests 35
4.2.1. Natural durability against fungi 35
4.2.2. Inhibition of fungal growth by extractives 36
4.2.3. Exposure to termites in laboratory 39
4.3. Dimensional stability 41
4.4. Chemical composition of Prunus africana 41
4.5. Scanning electron microscopy 42
4.6. Spectra analysis 44
4.6.1. 1H NMR analysis 44
4.6.2. Infrared analysis 46
5. Conclusion 48
Recommendations and prospects 48
Part 2. Improvement of Grevillea robusta durabilily 49
Introduction 49
Part 2. Chapter 1. Impregnability of Grevillea robusta using sap
displacement method 54
1. Introduction 54
2. Materials and methods 56
2.1. Sampling 56
2.2. Determination of fungal susceptibility of Grevillea robusta 56
2.3. Sap-displacement treatment 57
3. Results and discussion 59
3.1. Fungal susceptibility 59
3.2. Preservative retention 60
3.3. Preservative Penetration 62
4. Conclusion 66
Part 2. Chapter 2. Evaluation of thermally modified Grevillea robusta
heartwood as an alternative to shortage of wood resource in Kenya :
characterisation of physicochemical properties and improvement of
bio-resistance 67
1. Introduction 67
2. Materials and methods 70
2.1. Materials 70
2.2. Heat treatment 70
2.3. Exposure to fungi 71
2.4. Exposure to termites 72
2.5. Determination of the amount of extractive 73
2.6. Spectroscopic analysis 73
2.6.1. FTIR analysis 73
2.6.2. CP/MAS 13C NMR analysis 74
2.6.3. 1H NMR analysis 74
2.7. Contact angles measurements 74
2.8. Microscopic analysis² 74
2.9. Determination of Grevillea robusta chemical composition 75
2.9.1. Determination of lignin content 75
2.9.2. Determination of holocellulose content 75
2.9.3. HPLC analysis 75
2.9.4. Acidity titration 76
2.10. Mechanical strength test (MOR and MOE) 76
3. Results and discussion 77
3.1 . Heat treatment 77
3.2. Improvement of durability 78
3.2.1. Improvement of durability to fungi 78
3.2.1.1. Grevillea robusta natural durability 78
3.2.1.2. Grevillea robusta conferred durability 79
3.2.2. Improvement of durability to termites 84
3.2.2.1 Heat-treated Grevillea robusta against termites in the laboratory 84
3.2.2.2. Heat-treated Grevillea robusta against termites in the field 85
3.3. Spectoscopic analysis 86
3.3.1. CP/MAS 13C NMR Analysis 86
3.3.2. FTIR Analysis 91
3.4. Chemical composition of Grevillea robusta wood 92
3.4.1. Extractives content 92
3.4.2. Lignin and holocellulose content 92
3.4.3. Acidity titration 93
3.4.4. Sugar composition 95
3.5. Microscopic analysis 95
3.6. Mechanical properties 97
4. Conclusion 101
General conclusion and recommendations 102
Conclusion générale et perspectives 104
References 107
Appendix 118
1
Introduction générale
Ce travail a été réalisé en collaboration entre le Laboratoire d’Etudes et de Recherches
sur le Matériau Bois (LERMAB) à l’Université Henri Poincaré Nancy 1 (France) et le
Département des Sciences et Technologies du Bois à l’Université Moi (Eldoret , Kenya)
Le travail porte sur l’étude de différentes essences de bois indigènes du Kenya, comme
le Prunus africana qui est une essence naturellement très durable et le Grevillea robusta qui,
au contraire du précédent, est peu durable et très sensibles aux champignons, aux insectes et
aux termites.
La durabilité du bois varie d’une essence à l’autre dictant ses possibilité d’utilisation.
Normalement, les essences durables sont utilisées pour des conditions extérieures correspondant
à des risques de dégradation importante, alors que les essences moins durables sont utilisées pour
des applications intérieures ou les risques de dégradation sont plus faibles. Le traitement du bois
par des produits de préservation est coûteux dépendant de la méthode et des produits utilisés,
mais peut être économique à long terme en augmentant la durée de vie en service d’essences peu
durables.
Au Kenya, il existe de nombreuses essences de bois utilisées à des fins différentes en
fonction de leur abondance et de leur durabilité. La plupart des essences durables telles que
Prunus africana, Juniperus procera, Olea europea, Elgon teak, Meru oak, African mahogany ont
été surexploitées. Différentes raisons sont à l’origine de cette surexploitation : l’augmentation de
la population, la non prise en compte d’une politique de gestion durable des ressources
forestières, et enfin les faibles rendements matière lors de l’exploitation du bois conduisant à un
gaspillage important particulièrement dans les scieries.
L’industrie consomme également une quantité importante de bois provenant de forêts
naturelles gouvernementales telles que Elgon teak, Prunus africana, African mahogany,
Juniperus procera ou de plantations Pinus patula, Pinus radiata, Cupressus lusitanica, Eucalypts
etc. Tous ces facteurs, causent d’importantes pressions sur la forêt du fait d’une consommation
2
plus importante que sa régénération. En 1999, le gouvernement Kenyan a mis en place des lois
interdisant l’exploitation sauvage des forêts gouvernementales et intensifiant la replantation. La
conséquence de cette politique a été une pénurie importante de bois dans tout le pays conduisant
à l’utilisation d’essences moins durable peu utilisées jusqu’alors, comme le Grevillea robusta,
essence agro-forestière à croissance rapide très répandue au Kenya en particulier dans les régions
agricoles. Ce bois peu durable est actuellement utilisé en utilisation intérieure sous forme de
poutres et pour des applications extérieures telles des poteaux ou des clôtures. Cependant, de part
sa situation géographique, les conditions climatiques du Kenya sont propices à la dégradation du
bois par les termites, les insectes et les champignons conduisant à présent à un autre problème
concernant le remplacement des bois en service après des périodes relativement courtes de 2 à 3
ans. Des estimations montrent que si la situation perdure telle qu’elle est aujourd’hui, la pénurie
de bois augmentera dans les années à venir du fait de la faible qualité en service des bois à
croissance rapide utilisé.
Le gouvernement Kenyan, au travers du Ministère des Ressources Naturelles, a chargé
l’Université Moi de mettre en place un programme de sensibilisation destiné à la population
locale sur l’utilisation du bois. Un des objectifs est de mettre en place une information sur des
techniques de préservation peu coûteuses et faciles à mettre en oeuvre du fait de la difficulté
d’avoir recours à des techniques plus industrielles, pour des raisons de coûts principalement, tels
que les imprégnations vide-pression impliquant des produits comme la créosote ou les CCA. De
plus, il est important de fournir des informations sur des méthodes de traitement plus
respectueuses de l’environnement du fait de l’interdiction progressive de différents biocides dont
la créosote et les CCA étant donné leur toxicité. Ce programme, développé à l’Université Moi, a
été supporté par le gouvernement français grâce à une bourse d’étude attribuée par l’Ambassade
de France à Nairobi.
La première partie de ce travail concerne l’étude du Prunus africana dans le but de mieux
connaître les raisons de son exceptionnelle durabilité naturelle. Cette essence est utilisée pour des
applications extérieures aux contact du sol et est résistante aux termites et aux champignons. Des
traverses de ce bois ont été utilisées pour les voies de chemin de fer au Kenya et en Uganda
depuis plus de 90 ans et sont toujours en service. Les extractibles présents dans l’écorce
3
possèdent des propriétés biologiques intéressantes et sont utilisés depuis plus trente années en
médecine traditionnelle pour soigner des cancers de la prostate. Malgré ces propriétés
intéressantes des extractibles de l’écorce, aucune étude n’a été rapportée sur les extractibles du
bois de cœur et leurs effets sur sa durabilité naturelle. Plusieurs études décrivent la possibilité de
développer de nouveaux biocides en se basant sur des analogies avec des produits présents dans
les extractibles de bois naturellement durables (Inamori et al., 2000 ; Baya et al., 2001). De tels
produits pourraient, après identification et synthèse chimique, constituer des composés moins
toxiques pour l’homme et plus respectueux de l’environnement. La valorisation d’extraits
naturels pourrait également trouver des applications dans le domaine de la cosmétique dans des
produits comme des crèmes lavantes ou des shampoing ou des propriétés biocides légères sont
requises.
La seconde partie du travail concerne l’augmentation de la durabilité du Grevillea robusta, qui
est une essence très sensible à la dégradation lorsqu’elle est placée en conditions extérieures. Son
imprégnabilité par des solutions de CCA a été étudiée sur des poteaux provenant d’arbres
fraîchement abattus en utilisant une méthode de déplacement de sève. Cette méthode constitue
une alternative intéressante facilement utilisable pour les populations locales pour traiter sur
place des poteaux destinés à des clôtures ou palissades. Les taux importants de pénétration et
rétention de la solution de CCA permettent d’envisager des utilisation en classe de risque 4 au
contact du sol ou les attaques par les insectes, les termites et les champignons sont importantes.
Le traitement thermique, méthode récente de préservation du bois n’impliquant pas
l’imprégnation de biocides à l’intérieur du bois, a été appliquée au cas du Grevillea robusta. Des
travaux précédents montrent que ce type de traitement permet d’augmenter la durabilité du bois
vis à vis des champignons de pourritures. Appliqué au cas du traitement du Grevillea robusta, ce
traitement pourrait trouver des applications dans le cas de la fabrication de meubles, de murs,
plafonds, toitures et autres clôtures, où le matériau n’est pas en contact direct avec le sol. Une
telle méthode présenterait également l’avantage d’être plus respectueuse de l’environnement, ce
qui est un des impératifs dans le domaine de la préservation du bois.
4
General introduction
The research was conducted in collaboration between the LERMAB at University
Henri Poincaré, Nancy-1 (France) and the Department of Wood Science and Technology at
Moi University (Kenya).
The study involved research on Kenyan wood species Prunus africana that is naturally
durable and Grevillea robusta, which is not durable and very susceptible to attack by fungi,
termites and woodborers. Natural durability of wood varies from one species to another. The
quality of wood in its natural form dictates the end use. Normally, durable wood species are
utilized externally where the risk of degradation is high while less durable woods are used for
internal construction purposes where the risks of degradation are lower. Chemical treatment of
wood is expensive depending on the method of treatment but it’s economical in the long run by
increasing the service life of less durable wood.
In Kenya there are many different wood species that are utilized for different purposes
depending on the durability and availability. Most of the durable wood species such as Prunus
africana, Juniperus procera, Olea europea, Elgon teak, Meru oak, African mahogany have been
over-exploited from the government forests. This was due to increase in population, lack of
proper knowledge on sustainability of forest resources and poor recovery during the conversion
process leading to wastage mainly by saw millers. Wood based industries have also been logging
from the government forests for indigenous species eg Elgon teak, Prunus africana, African
mahogany, Juniperus procera and plantation exotic wood species such as Pinus patula, Pinus
radiata, Cupressus lusitanica, Eucalypts etc. This caused a lot of pressure to the forest resources
because the rate of logging was higher than replanting. In 1999, the Kenyan government effected
logging ban from the government owned forests and intensified replanting exercise in all forest
stations. This caused a serious wood shortage in the whole country. People resulted to using other
alternative sources especially the fast-growing agro-forest tree species like Grevillea robusta.
This species is widely spread in Kenya mostly in the agricultural prime regions. Unfortunately, it
is not durable and due to acute shortage of wood, it is being used for construction timber and
externally as fencing posts. Since Kenya lies in the tropics, the temperatures are generally high
5
and degradation of wood by termites, insects and fungi is quite high (Bagine 1992). This leads
yet to another problem of replacing low durable wood in use after short time of about 2 to 3
years. Projection reports show that at the present rate of utilization, wood shortage will be acute
in the near future due to short service life offered by fast growing species.
The Kenyan government, through the Ministry of Natural Resources requested Moi
University to provide information on wood durability in relation to end-use by the local people.
Also to advice the local people on cheap wood treatment methods that are applicable on-site
because the available Chromated Copper Arsenate (CCA) and creosote pressure treatments are
expensive and out of reach for local people. To add on this, it was important to give information
on modern technologies on wood treatment that are environmentally friendly because some
preservative chemicals such as CCA are being phased out due to their negative effects to the
environment. Financial assistance was provided by the French government through the French
Embassy in Nairobi, under staff development training programme in Moi University, Kenya.
The first part of my work involved working on Prunus africana wood species from Kenya
to have better understanding of its natural durability. This species has been used for exterior
purposes and is resistant to attack by termites, insects and fungi. Bulk of this wood was used for
construction of Kenya-Uganda railway more than 90 years ago with minimal treatment and is still
in place. Extracts from Prunus africana bark have very interesting medicinal properties. For more
than 30 years now, Prunus africana bark has been used traditionally for the treatment of prostrate
cancer. Since the extractives from the woody parts of Prunus africana have never been analyzed,
it was important to carry out the research to have more understanding on wood durability.
Biological resistance against termites and fungi gave lead to durability of this species due to
presence of extractives. Several studies have reported the possibility of developing new biocides
analogous to natural products present in extractive matter (Inamori et al., 2000 ; Baya et al.,
2001). Such products if developed synthetically would be more environmentally acceptable due
to their composition based on analogies with natural extracts present in wood. Valorization of
durable wood extracts as additives in other products, which are required to have mild fungicidal
properties such as cosmetics, soaps, shampoo form could also be an interesting subject.
6
The second part of the work was to improve the durability of Grevillea robusta, which is
very susceptible to attack after short time in service especially for external use. Sap displacement
treatment method using CCA was investigated using freshly cut Grevillea robusta trees. This is a
cheap treatment method with high potential of being used by local people to treat fencing posts
on site by use of available apparatus such as plastic containers. Effective impregnation using
CCA makes it possible to use Grevillea robusta in high hazard areas infested with aggressive
wood degraders like termites, fungi and woodborers. Heat treatment is a modern technique of
wood preservation that does not involve impregnation of chemicals into the wood. This method
was applied in the treatment of Grevillea robusta. Previous work done on heat treatment shows
that this form of treatment improves durability of wood against fungi. Such treatment would be
ideal for Grevillea robusta used for furniture, wall, ceiling, roofing, flooring and fencing
purposes. Also the method is environmental friendly, which is a requirement in the wood
preservation industry.
7
Part 1. Effect of extractives on natural durability
of Prunus africana wood species
1. Context of study
1.1. Natural durability of wood
Due to its durability and workability using simple tools, wood has long history in use
dating back more than 4000 BC for construction of boats and houses. Natural durability of wood
is its ability to resist biological degradation without any form of treatment. Generally, wood
durability is categorized into three categories :
- very durable wood that survives more than 15 years in service,
- moderate durable 10-15 years,
- non-durable has service life of less than 10 years.
Durability is the main factor considered in the utilization of wood. Some wood species
have very high durability and endure for long time in service even in areas prone to attack by
termites, insects or fungi. Wood extractives play an important role in the natural durability
(Reyes Chilpa et al., 1998 ; Celimene et al., 1999 ; Mori et al., 1997 ; Haupt et al., 2003 ;
Windeisen et al., 2002 , Neya et al., 2004 ; Neya, 2002 ; Gérardin et al., 2004). According to the
wood species and to the nature of extraction solvent, extractives present different inhibition rates
for fungi.
In Europe, wood was a dominant building material dating back to the beginning of
civilization. Structural wood was placed directly in the ground exposing it to aggressive
degraders but still the structures lasted for many years. The most famous ancient European wood
buildings still in evidence today are Norwegian stave churches, hundreds of which were built in
the 12th and 13
th centuries of which about 30 are still surviving with their original wood structure.
In North America wood was in abundance and this led to widespread of its use for
construction of small structures. The oldest surviving homes in the US date back to early 1600,
8
while about 80 homes of this period are still surviving in the New England states. Even in
Louisiana, where the climate is hot and humid presenting challenge to wood durability, some
original French settlements are found dating back to the first half of 1700s. There are many more
wood buildings from 1800s and early 1900s still in good condition and occupied.
Japan has a well-known history of wood use and is the home of the oldest surviving wood
structure in the world, a Buddhist temple near the ancient capital city of Nara. The Horyu-ji
temple is believed to have been built at the beginning of 8th century or even before. The longevity
is due to good maintenance and repair.
Apart from durability, wood has more advantages as compared to other materials. Ease of
construction, thermal performance and aesthetics. In addition there are many specific applications
where the natural durability properties make wood a material of choice. Being resistant to
chemical destruction, it’s often the material of choice when exposed to acids, salts and organic
compounds. Throughout the world, wood bridges have proved to be remarkably durable. Modern
bridge decks are subjected to relentless attack of de-icing chemicals and wood is being accepted
as a viable option of these applications. Pilings that are constantly submerged in fresh water last
for centuries. Foundation piles under structures never decay if the water table remains higher than
the pile tops. Many of the world’s important structures are built on woodpiles including the city
of Venice and the Empire State Building in New York.
1.2. What is known about Prunus africana
Prunus africana (figure 1) is also known as Pygeum africanum and commonly as red
stinkwood native to the Afromontane forests of Africa (Walter and Rakotonirina, 1995). The
species is naturally durable against most of the wood biodeterioration agents. In Kenya, it is
mainly used for medicinal purposes and construction, mostly in areas prone to attack by
woodborers, termites and fungi. Depletion of this species has raised concern because even after
the government ban on logging, illegal exploitation is still evident in some parts of the country
(Hitimana, 2000). Local farmers are encouraged to domesticate some high value tree species like
9
Prunus africana to ensure their continued existence. This involves accelerated and human-
induced evolution to bring species to wider cultivation through farmer-driven process.
Figure 1. Prunus africana
Typically, mature Prunus africana tree is about 25m tall and diameter of upto 1m
(ICRAF, 1992). Individual trees as large as 1.1m to 1.5m dbh has been recorded in Mt Elgon
lower Afromontane forest, Kenya and Malawi (Hitimana, 2000). In forest condition, the bole is
straight, cylindrical and sometimes it may be free of branches for upto 20m or more (Letouzey,
1978). It is evergreen species with simple and alternate leaves with small green-white flowers in
short spray. The fruits are round, dark red containing one seed. The species is fairly slow growing
and it’s propagated through seedling and wilding (ICRAF, 1992).
The medicinal value of its bark was recognized more than 30 years ago involving
commercial harvesting of the species from natural populations in some African countries. It is a
10
traditional medicine in Africa used to treat chest pain, malaria and fever (Cunningham and
Mbenkum, 1993 ; Kokwaro, 1976). On the international market, it is traded for manufacture of
products used to treat prostrate gland hypertrophy and closely related but more serious condition
of benign prostatic hyperplasia. The major exporters of the bark include Cameroon, Madagascar,
Equatorial Guinea and Kenya to Groupe Fournier of France and Indena of Italy that produce 86%
of the world’s bark extract (Stewart et al., 2003). Prunus africana is traded in the form of dried
back. About 2000 kg of fresh bark representing 1000 kg of dried bark are needed to make 5 kg of
extract (Cunningham et al., 1997). Global demand for Prunus africana and its extracts rose from
2.4 million kg in 1995 to more than 2.8 million kg in 1996 and again to 3.1 million kg in 1997
(Cunningham and Mbenkum, 1993).
International trade for Prunus africana has been regulated by the Convention on Trade in
Endangered Species (CITES). Monitoring the trade of this species is difficult because its traded
in five different forms – unprocessed dried bark, bark extract, herbal preparations in form of
capsules, as constituent of hair tonic and as wood for manufacture of high quality furniture
(ICRAF, 1999 ; ICRAF, 2000). Despite the ban on cutting of this species, large quantities are
illegally harvested. Commercial exploitation, habitat loss and unsustainable harvesting have led
to its decline threatening conservation of its genetic diversity. The species has been listed as
venerable in the world list of threatened trees, owing to its rapid population declines. This has
been fuelled by unsustainable extraction methods involving excessive debarking or felling the
entire tree due to high demand (ICRAF, 1999).
The extracts from Prunus africana contain several pharmacologically active compounds
that interfere with the development of benign prostatic hyperplasia (BHP). Phytosterols (β-
sitosterol, β-sitostenone) reportedly inhibit the production of prostaglandins in the prostrate,
which suppresses the inflammation symptoms associated with BPH and chronic prostatis (Breza
et al., 1998 ; Stewart et al., 2003 ; Catalano et al., 1984). Pentacyclic triterpenes (oleanolic and
ursolic acids) are believed to inhibit the activity of glucosyl-transferase, an enzyme involved in
the inflammation process. Ferulic esters lower the blood levels of cholesterol, from which
testosterone is produced. Many compounds have been identified including fatty acids (C12–C24),
sterols (β-sitosterol, β-sitosterol-3-O-glucoside, β-sitostenone and campesterol), pentacyclic
11
triterpenoids (ursolic acid, 2-hydroxyursolic acid, oleanolic acid, crataegolic acid, maslinic acid,
epimaslinic acid, friedelin) and two linear alcohols, n-tetracosanol and n-docosanol and their
trans-ferulic esters. The efficacy of the extracts in the treatment of benign prostate enlargement in
man has been confirmed in clinical and pharmacological experiments (Thieblot et al., 1971 ;
Thieblot et al., 1977 ; Dufour et al., 1984 ; Bassi et al., 1987 ; Barlet et al., 1990 ; Bombardelli
and Morazzoni, 1997). These compounds and others are believed to work synergistically to
counter the structural and biochemical changes associated with the disease. Studies have shown
effectiveness and safety of Prunus africana bark extracts in the treatment of BHP symptoms
including frequent urination, painful urination and urgency, producing only minor side effects
(Stewart et al., 2003, ICRAF, 1992).
Main components identified in chloroform extracts of Prunus africana bark are presented
in figure 2.
Prunus africana as a construction wood is durable against fungi and termites in Kenya.
Wood may have variation in resistance against biodegradation even within species. This variation
is attributed to genetic difference in wood (Hart, 1982).
12
Figure 2. Main components contained in extracts of Prunus Africana bark
COOHH
H
HHO
Oleanolic acid
COOHH
H
HHO
Maslinic acid
HO
CH3(CH2)21OH
n-Docosanol
CH3(CH2)23OH
n-Tétracosanol
HH
Friedelin
O
HO
H
H
iPr
Et
β-Sitostérol
COOH
H
HHO
Ursolic acid
COOHR1
R2
OOC-CH=CH
OCH3
OH
Fatty acids among which palmitic acid
CH3(CH2)14COOH
trans-ferulic acid esters
13
2. Literature review
2.1. Wood biodegradation
Wood is an organic material mainly composed of cellulose, hemicelluloses and lignin. These
are distributed in the cell wall in different proportions among wood species. The ray parenchyma
cells of some wood species contain starch, soluble sugars, proteins and extractives (Freas, 1982 ;
Schultz et al., 1995). Biological deterioration of wood has been of concern to the timber industry
due to the economic losses caused to wood in service or in storage. Biodeteriogens of wood include
fungi, insects, termites, marine borers and bacteria. These are conventionally referred to in terms of
the kind of organisms responsible for particular damage (Nicholas, 1973 ; Cartwright and Findlay,
1958).
Different wood degraders attack different components of the wood at different speeds (Scott,
1968). Degradation helps to maintain the balance of nature by returning the degraded material into
the soil to sustain fertility of soil and growth of new plants. Attack on wood by the degraders
depends on the environmental conditions, whether in storage or in use (Zabel and Morrel, 1992).
Fungi that degrade wood reproduce largely by spores, which are in simple structure and of
microscopic size. They produce thread like hyphae that penetrate into the wood cells. Germination
of spores of decay fungi may be favoured not only by appropriate moisture and temperature, but also
by acidic conditions and readily utilizable nutrients especially sugars at the wood surface and a
certain amount of carbon dioxide. Fungi have the ability of detoxifying toxic chemicals in wood
especially preservatives and extractives (Schultz and Nicholas, 2000). This means that they can
cause extensive damage in treated and untreated wood more so on softwoods. They hydrolyze wood
components and assimilate them as food by injecting enzymes into the wood cells (Eriksson et al.,
1990). When wood is maintained at elevated temperatures for long periods, it is susceptible to
deterioration for example wood chips in storage before processing. A microflora composed of
thermophilic and thermo-tolerant microorganisms becomes established throughout the pile and fungi
isolated from different positions reflect the temperature gradient from the centre (Hale and Eaton,
1986). Fungi are by far cause of the greatest economic losses to timber through hydrolysis of the
cell wall components. They produce diffusible enzymes that catalyze the dissolution of the
14
components and this causes progressive changes of properties. The main groups of wood destroying
fungi include soft rot, white rot and brown rot fungi (Anke et al., 2006). Toughness is a mechanical
property, which is most sensitive to wood deterioration. A decrease in toughness is noted even
before weight loss is detected. Other mechanical properties affected as degradation progresses
include bending strength, compression, hardness and elasticity (Schirp and Wolcott, 2005).
2.2. Chemical nature of extractives associated with wood durability
The chemical contents vary from species to species but generally, hardwoods have higher
tannin contents than softwoods (Pizzi et al., 1986). Schultz and Nicholas (2000), suggested that
extractives have the ability to protect wood by combination of fungicidal and antioxidant
properties. Also phenolic extractives have the ability to complex with metals i.e. extractives are
metal chelators and this is an additional means of wood protection by extractives (Hillis and
Sumimoto, 1989 ; Slabbert, 1992). Wood degradation by fungi often involves various metals, either
in free form or as key components of enzymes (Ericksson et al., 1990 ; Green et al., 1997).
Extractives are composed of different components that could be extracted from wood using
various solvents. According to their chemical structure and abundance extractives could be more or
less active against wood's biodegraders leading to different class of natural durability for wood.
Some compounds isolated from wood extractives and identified as responsible of wood natural
durability are present thereafter.
- Flavonoids
In 1995, Schultz et al. attributed the durability of acacia to the presence of dihydromorine
and dihydrokaempferol (figure 3). Dihydromorine was also noted to contribute in the durability
of Morus mesozygia (Déon et al., 1980).
15
OHO
OH
OH
HO OH
O
OHO
OH O
OH
OH
Dihydromorine Di-hydrokaempferol
Figure 3. Structure of flavonoids responsible of the durability of accacias
Reyes chilpa et al. (1998) worked on plalymiscium yucatanum and noted that there are up
to five components which are responsible for the durability against fungi among which 4,2,5-
trihydroxychalcone was identified as particularly active against brown rots and white rots
(figure 4).
OH
OH
OH
HO
Figure 4. Structure of 4,2,5-trihydroxychalcone
The natural durability of wood bark is due to the presence of many substances like D-
gallocatechine, D-catechine and also many condensed tannins. The role of flavanoids in
biodegradation resistance is variable and dependent on the test fungi. Robinia pseudoacacia pose
exceptional resistance to biodegradation, attributed to the concentration of robinetine and
dihydrobine (Rudman, 1963). Wood preservative formulations based on tannin/copper,
tannin/zinc, tannin/boron, have been proposed (Thévenon et al., 2001). Wood preservatives based
on tannin/copper are very effective against fungi and insects (Toussaint, 1997).
- Stilbenes
Research has shown that stilbenes are responsible for resistance against fungi even in
vegetables (Adaskaveg, 1992). The 2,4,3’,5’-tetra and 3,4,5,3’,5’-penta-hydroxystilbenes (figure 5)
16
have been found to be responsible for resistance against brown rot fungi (Schultz et al., 1995).
Stilbene phytoalexins can be modified in vivo by partial methylation to enhance bioactivity (Schultz
et al., 1997).
HO OH
OH
OH
OH
HO
OH
OH
OH
2,4,3’,5’-tetrahydroxystilbene 3,4,5,3’,5’-pentahydroxystilbene
Figure 5. Structure of some stilbenes isolated from wood
Stilbenes (pinosylvine, pinosylvin monoethylether and pinosylvine dimethylether) are known
to control the activity of brown and white rot fungi (Schultz et al., 1997).
- Quinones
Wood and vegetable material contain many different quinoid structures. Tectona grandis
wood contain about 0.3% of 2-methylanthraquinone (tectoquinone) and also other quinones like
naphtoquinone (Thévenon et al., 2001). Tectoquinone is assumed to be at the origin of the resistance
of teck wood to termites (Levya et al., 1998).
CH3
O
O
OH
O
O
Tectoquinone Naphtoquinone
Figure 6. Structure of some quinoids
17
- Lignans
Lignans are intermediates in the biosynthesis of lignin derived from oxidative coupling of
coniferylic or coumarylic alcohols. Some of them have been identified as toxic to biodegraders and
difficult to decompose (Pandey and Pitman, 2004). Different structures have been identified in
various wood species as shown in Fig. 7. Lignans present in the bark of Obvata sp are particularly
important in controlling activities of basidiomycetes (Mori et al., 1997).
O
OH3CO
HOC3H7O
OCH3
OH
Figure 7. Structure of 7-isopropoxymatairesinol
- Terpenes
Monoterpenes like α and β-pinenes are toxic to Heterobasidon annosus (Flodin and Fries,
1978). Mixtures of monoterpenes like limonene, myrcene, 3-carene, α and β-pinene, which are
contained in oleoresin of Pinus ponderosa, inhibit growth of fungi and bacteria.
Limonene Myrcene ∆3-Carene
α-Pinene β-pinene
Figure 8. Structure of different monoterpenes
High levels of wood extractives were associated with the exceptional durability of Prosopis
africana against fungi and insects (Gérardin et al., 2004). The strong hydrophobic character of wood
and high dimensional stability could also be significant factors contributing to wood resistance
(Neya et al., 2004).
18
Wood traits with ether extract at 4% showed high resistant against Caniophora puteana,
Chaetonium globosum and Gloephylum trabeum (Micales et al., 1994). Extracts from Juniperus
procera indicated presence of ten distinct components, which were found to have anti-termite effects
(Kinyanjui et al., 2000). Extracts from red cedar sawdust showed increased resistance from attack by
Gloephyllum trabeum fungi at 0.2% p/v (Lopez-Carbonell et al., 1998).
2.3. Wood attacking microorganisms
Microorganisms have always caused problems to wooden materials. Fungi and bacteria
attack both living and dead trees in different ways. The organisms have different demands in
terms of conditions that sustain life. These conditions are specific to the species of organism and
substrate they feed on (Morton and Eggins, 1976). However, all living organisms need water and
nutrition. At very high moisture content values, wood degrading fungi are not able to grow, while
bacteria survive.
Fungi are members of the plant kingdom but differ from other plants by not processing the
green colouring matter, chlorophyll. Despite the great diversity in form of fungal fruit bodies, the life
cycle pattern in the majority of filamentous fungi has many similarities (figure 9).
Figure 9. Life cycle of wood destroying fungi
19
The asexual part of the life cycle involves production of large number of spores, which are
disseminated into the environment. These germinate when the environmental conditions are suitable
to support mycelia colony. These include temperature, pH and moisture levels. Formation of sexual
reproduction is associated with the onset of adverse climatic conditions. This involves the fusion of
two differentiated cells followed by fusion or karyogamy and then meiosis, which result in
segregation of genetic characters in progeny. Fungi have a requirement for organic compounds as a
source of energy and carbon, which are used in normal cell metabolism. Organic and inorganic
compounds are available to fungi in form of cellulose, hemicellulose, lignin, soluble sugars and
minerals (Presnell and Nicholas, 1990). Different families of fungi are susceptible to colonize and
degrade wood.
2.3.1. Moulds
Members of ascomycetes and fungi imperfecti cause molds and stains on wood. These
colonize freshly felled wood without causing significant damage to cell walls. The presence of water
in wood is an essential requirement for fungal colonization and decay to occur. Wood that attains
moisture content of 20% and above is susceptible to attack (Eaton and Hale, 1993).
They cause discolouration by pigments within the penetrating hyphae or by pigments only in
the surface-formed conidia (in case of stains). Moulds cause little damage to the structure of wood
they inhabit, provided their action does not reach a more aggressive stage where it would be
considered soft rot (Schultz et al., 1995).
Moulds are always present in our surroundings as airborne spores. They do not have any
of their own enzymes for degrading wood, and only live on the surface of the wood (Fengel and
Wegener, 1984). They are more dependent on the humidity in the air and on the surface than on
the moisture content in the substrate (Assersson and Bergman, 1971). The lowest relative
humidity, RH that can support mould growth on pine and spruce has been shown to be 80%
(Viitanen and Ritschkoff, 1991a). The more hygroscopic a material is, the lower the relative
humidity required for supporting mould growth (Block, 1953). Since moulds are a large group of
fungi, they demonstrate a large range of temperature tolerances. The mould fungi generally grow
20
in temperatures around 0-50˚C, while some species even propagate at temperatures down to
negative 6˚C. Mould that is present on a wooden surface can have both an antagonistic and
stimulating effect on other fungi (Fengel and Wegener, 1984). Furthermore, since they have an
ability to bind water, they cause higher moisture content near the surface, and consequently they
can make the wood more susceptible to other fungi. Therefore, a mould attack can be seen as an
indication of impending rot attacks.
2.3.2. Sapstain
The term "sapstain" is used as a cover term for different microfungi that attack and
discolour wood. Insects often disperse sapstain fungi. They also need free water in the wood, e.g.
moisture content (MC) above 25-30%, to survive. They grow at temperatures between 0-50˚C
and have different optimal temperatures, depending on species. Some species even grow at
temperatures a few degrees below zero (Käärik, 1980). Their hyphae contain a pigment (melanin)
of a blue, green or black colour, which causes a discoloration in the wood (Fengel and Wegener,
1984 ; Brisson et al., 1996). The pigment is not produced at lower temperatures. Consequently, a
seemingly unaffected wooden surface can be extensively discoloured as the temperature
increases. In contrast to moulds, the sapstain hyphae grow into the wood through the rays,
bordered pits and cell lumen (Fengel and Wegener, 1984). Although it grows in the cells, it
cannot feed on cellulose and therefore it does not affect the strength properties of the wood
(Blanchette et al., 1992).
2.3.3. Brown rot
Brown rot is caused by Basidomycota, a group of fungi with characteristic fruit bodies
and a sexual reproduction system (Deacon, 1997). These fungi are the most common rot type on
wood used above ground. The optimal MC in wood for brown rot growth is 30-70% (Viitanen
and Ritschkoff, 1991b). It mainly degrades cellulose and hemicelluloses in the cell wall, leaving
the wood to turn brown, caused by the presence of oxidised lignin (Green et al., 1997). A typical
cubical cracking is also observed in wood with brown rot. The fungi’s hyphae grow in the cell
lumen, in the same way as is observed in soft rot. However, in contrast to soft rot, brown rot
21
degrades almost the whole of the S2-layer of the cell wall (Deacon, 1997). There is no
localization of decay around the hyphae of the brown rot fungus. Instead, the fungus decays the
cell walls by means of a generalized thinning (Carlile and Watkinson, 1994). In the later stages of
brown rot decay, the cell wall collapse and small splits within the wall very close to the S2
microfibrillar orientation are visible. Severely brown rotten wood is brown in colour and brittle due
to the amorphous nature of lignin (Anke et al., 2006).
2.3.4. White rot
The white rot fungi include both Ascomycota and Basidomycota (Deacon, 1997).
Ascomycota is distinguished by a sexual reproduction in so-called ascus. White rot fungi are
capable of metabolizing all the major components of wood cell wall, both lignin and carbohydrate
fraction. It has also been shown that white rot fungi are effective in degrading various classes of
hydrophobic extractives found in Scots pine (Dorado et al., 2001). White rot do not only degrade
the S2-layer (as does soft rot), but also erodes the S1 and S3-layer sequentially. In the early
stages of an attack, each hypha is surrounded by an erosion zone or a bore-hole (Carlile and
Watkinson, 1994). In the advanced stages of white rot there is little or no shrinkage or collapse of
cells. The original shape and outward appearance of wood are maintained. White rotten wood
appears white and fibrous due to residual cellulose microfibrils. Severely attacked wood by white rot
has reduced mechanical properties (Nicholas, 1973).
2.3.5. Soft rot
Soft rot is caused by Deuteromycota and Ascomycota. Deuteromycota, propagate through
asexual spores, conidia. It decays very wet wood and mainly occurs in ground contact, since it
requires nitrogen. The nitrogen is taken up from the surrounding soil (Deacon, 1997). 73 of the 81
microfungi isolated from wood and tested have been found to cause soft rot decay (Mouzouras,
1989). This provides evidence that substantial number of soft rot can cause serious economic loss to
the pulp yield from wood chips by hydrolyzing cellulose during storage. It shows a preference to
degrade cellulose, and it causes characteristic cavities in the S2-layer of the cell wall. In the
initial colonization, it grows in the cell lumen, and from this position a lateral branch penetrates
22
the thin lignified S3-layer. As the hyphae enter the S2-layer, it branches into two hyphae and
grows up and down the cell wall in the direction of the microfibrils (Carlile and Watkinson,
1994). The soft rot fungi are found growing on wood in high moisture content (Anke et al., 2006).
The degradation occurs slowly in softwood when compared to white- or brown rot. Wood that
has been subjected to a soft rot attack looks similar to that attacked by brown rot, but has a
somewhat softer surface and reduced mechanical properties (Eriksson et al., 1990).
2.3.6. Bacteria
There are many different types of bacteria that feed on wood. Most of them are
Actinomycetes. Bacterial attacks are slower in comparison to attacks from most fungi. This is
often due to the fact that bacteria needs free water in the wood cell to propagate (Fengel and
Wegener, 1984). Bacterial degradation of the cell wall occurs by the action of tunnelling and
erosion bacteria that live in wood that is in contact with soil. In very wet wood out of ground
contact, for example logs stored in water, bacteria only degrade the pit structures (Clausen,
1996). In such attacks, bacteria invade the wood through the ray parenchyma cells and form
colonies in the sapwood where they feed on proteins (Fengel and Wegener, 1984).
Bacteria are normally the first colonizers of wood. They precondition the wood for invasion
by decay fungi by creating conditions necessary for attack both by altering the nutrient status of the
wood and by the production of synergistic secondary metabolites (Holt et al., 1979). Bacteria
breakdown pit membranes in the sapwood opening up the wood structure making fungal growth
possible. Some bacteria are capable of fixing atmospheric nitrogen for subsequent fungal
colonization of the wood (Dickinson and Levy, 1979). They also have the ability to detoxify and
degrade wood preservative elements including heavy metals such as Cu, Cr and As rendering them
harmless. Bacteria are more tolerant to high lignin and extractive contents in wood, high
preservative loadings and low levels of oxygen than fungi. They may be the sole agents of decay in
such conditions where other decay organisms are excluded. The three different forms of bacterial
decay are recognized as: erosion, tunnelling and cavitations (Singh and Kim, 1997).
23
2.4. Termites
Termites are social insects and exist in well-defined colonies under crowded conditions. The
size of colonies varies from a few thousand individuals to millions. Most of them require high
moisture and carbon dioxide levels and shun the light (Cartwright and Findlay, 1958 ; Kollman and
Cote, 1984 ; Hickin, 1975). Termites exist worldwide and may be found in latitudes 50o North and
50o South. They are the most dangerous pests of wood material since they occur in large numbers
and within short periods can cause heavy destruction by hydrolyzing the wood components with
enzymes (Theodore, 1973). Out of about 2000 termite’s species, only a few are of economic
importance. These include species of subterranean Rhinotermitidae, Coptotermes, heterotermes and
shedorhinotermes in the tropics while reticulitermes are found in subtropical and temperate regions
(Hickin, 1971).
Termites feed almost exclusively on vegetable materials although extensive damage is often
done to other materials in their efforts to find cellulose (Gitonga, 1995). Some live only in wood
and are able to digest cellulose with the aid of protozoa secreting cellulase, which break down
cellulose into simpler materials capable of being digested (Nunes and Dickson, 1996). The higher
termites, particularly macrotermes in Africa and odontermes in Indomalaya, construct nests in which
they bring in fungi to break down wood components into simpler digestible substances (Lee and
Wood, 1971). A study carried out by Peralta et al. (2004) which focused on wood consumption
rates of different forest species by termites under field conditions, did not find a strong
correlation between wood density and termite resistance. The authors did, however, acknowledge
the importance of wood hardness as a deterrent to termite damage, concluding that wood density
alone cannot be considered the single most important factor in determining termite resistance.
Generally, wood suffers severely from insect damage especially termites and other boring
insects. From a 10m by 200m belt transect marked in termite infested semi-arid area of Kajiado
Kenya, a total of 149 prey items including different wood species foraged by ants and termites
were recorded over a period of 3 years. Termites and ants make up to 67.5% and 29.3%
respectively of these items (Nyamasyo, 1995).
24
2.5. Other wood degraders
2.5.1. Insects
Certain wood destroying insects are found attacking trees, others prefer recently felled or
even seasoned wood. Among these, the most destructive insects are from the order coleoptera,
isoptera and hymenoptera all of them being widely spread in tropical countries.
The order coleoptera is the largest in the class insecta and the families, which are most
destructive to wood in storage, include anobiidae, lyctidae and bostrichidae, referred to as powder
post beetles. Also included are platypodidae and scolytidae, referred to as "Ambrosia beetles" or
pinhole borers. Powder post beetles constitute the most economically important group of wood
destroying insects and are only overshadowed by termites. Larval stage of these beetles can last up
to 5 years hence causes extensive damage (Hickin, 1975).
Lyctus beetles are often associated with sapwood and they derive their nourishment
primarily from the starch reserve stored in the parenchyma ray cells. The larvae bore small tunnels
approximately 1.5 mm in diameter and repeated attacks by the insects reduce wood into fine flour-
like powder. They invade wood with moisture content as high as 30% (Schultz et al., 2005).
Ambrosia beetles are associated with dead and dying wood, and are a characteristic of wood
in storage. They bore into the wood and feed on wood-inhabiting ectosymbiotic fungi introduced
into the gallery system by the beetles and are referred to as "ambrosia fungi". The fungi breakdown
cellulose and lignin, producing organic molecules which can be assimilated by the beetles and also
synthesize chemicals essential for beetle development (Collins and Weber, 1987).
2.5.2. Marine borers
These are important degraders of wood in marine environments. They cause economic losses
on marine pilings and other structures along the coastal waters. They can be categorized into two
families, molluscans and crustaceans. Molluscans include teredo (shipworm), pholads and bankia.
Teredos are filter feeders and they go into the wood for food and shelter. They can be identified by
25
their worm-like bodies, which sometimes extend upto 2m and calcified tunnels, which they live in.
They have two syphons, one used for sucking water, which contain food nutrients and the other for
pumping out water from the body. They cause extensive damage to the wood by deep tunnelling,
leaving behind a small diameter hole on the surface that may not show the extent of damage inside
the wood (Johnson et al., 1973). The crustaceans include limnoria, sphaeroma and chelura. These
are crab-like creatures, which have mandibles and have the ability of moving hence they are not
restricted to attacking the same piece of wood. They are 3-4 mm long and attack wood by boring
shallowly about 1 cm below the surface. Repeated attacks on the same wood cause an hourglass
appearance on pilings, which develops a weak point in intertidal zone (McCoy, 1965).
26
3. Materials and methods
3.1. Wood species utilized
Prunus africana, which is naturally durable wood species from Kenya, was used for the first
part of this work. Sampling was done from trees aged 38 years at the time of cutting from Kapsabet
Forest Station, Kenya. Samples were cut from heartwood and sapwood for comparison purposes.
They were packed and transported to Nancy (France). In the laboratory, this wood was cut into
various sizes for different tests as stipulated in the methods.
3.2. Measurement of wood extractives
Heartwood and sapwood samples were cut into small blocks of 5 x 20 x 25 mm using a band
saw. These were kept at room temperature before grinding and soxhlet extraction
3.2.1. Soxhlet extraction
Prunus africana sapwood and heartwood were separately ground to fine powder, passed
through a 115-mesh sieve and dried at 60°C to constant weight before extraction. The temperature
of 60°C was chosen instead of the normal 103°C to avoid degradation of the extracts. Different
solvents including hexane, acetone, dichloromethane, water and mixture of toluene/ethanol at ratio
of 2:1 (v/v) were used for soxhlet extraction..
Two procedures were used for soxhlet extraction.
- Procedure a
Sample extraction was done using hexane, or dichloromethane, or acetone, or
toluene/ethanol or water. For each extraction, 10 g samples of sawdust were used. Extraction by
each solvent was done for 15 hours at a rate of about 10-12 cycles per hour. Three replicate
extractions were done for each sample.
27
- Procedure b (series extraction)
Four solvents : hexane, dichloromethane, acetone, toluene/ethanol were used successively
for extraction from the same sample of 80 g in the same order each for 15 hours.
3.2.2. Quantification of extractive
Solvent was removed under reduced pressure by Rotavapor and the residue dried over P2O5
under vacuum in a desiccator before weighing (me). The percentage of extractive was determined
by two methods.
- Direct measurement (DM)
The percentage of extractives was obtained from the weighed mass of extracts (me) obtained
after solvent evaporation according to the formula :
% 100xm
mDM
s
e=
where ms is the dried mass of the sawdust before extraction.
- Indirect Measurement (IM)
The percentage of extractives was obtained from the mass of extracts estimated by difference
between the dried mass of the sawdust (ms) before extraction and after extraction (mse) according to
the formula :
% 100xm
)mm(IM
s
ses −=
3.3. Biological tests
3.3.1. Prunus africana natural durability
In this test, correctness of theoretical dry mass of wood is assumed because for natural
durability test, samples should not be dried at 103°C to avoid degradation of the wood components.
28
Prunus africana heartwood samples were cut into 25 x 25 x 5 mm longitudinal, tangential and radial
respectively. These were conditioned at 22°C and relative humidity of 60-70% until they achieved
constant weight.
Theoretical dry mass (mt) of test samples was determined by calculating the average percent
(%) moisture content of similar samples dried at 103°C to get µ.
µ was the average humidity percentage measured after oven drying at 103°C.
µ) (100
m 100m u
t +=
where mt is theoretical dry mass of the test samples
mu is mass of test samples after conditioning at room temperature
µ is percentage average humidity
Petri dishes (9 cm diameter) were filled with sterile culture medium prepared from malt
(20 g) and agar (40 g) in one litre of distilled water, inoculated with Coriolus versicolor and
incubated at 22°C and 70% HR to allow colonization of the medium by the mycelium.
Two wood blocks sterilized by UV light were placed in the Petri dishes containing fully-
grown fungi. All experiments were triplicate. Beech (Fagus sylvatica) was used as control. Blocks
were exposed to fungus for 16 weeks at 22°C and relative humidity of 70%. Assessment was done
by determination of weight loss using the following formula :
Weight loss (%) = 100xm
)mm(
t
ft −
where mt is the theoretical dried weight of test samples and before attack
mf is the final ovendried weight of the sample after attack.
In the case of ovendried wood blocks, weight losses are calculated as follow :
Weight loss (%) = 100xm
)mm(
d
fd −
where md is the ovendried weight of the test samples before attack and mf the final ovendried weight
after attack.
29
3.3.2. Growth inhibition
Mycelium was grown in 9 cm Petri dishes filled with 20 ml of malt-agar medium
prepared by mixing 20 g of malt and 40 g of agar in 1 l of distilled water containing 100 or
500 ppm of extract. Introduction of the extracts was carried out after medium sterilization (20
min., 120°C, 1 bar) by addition of the necessary quantity of extract solubilized in the minimum
amount of ethanol. Plates were inoculated by placing a small portion of a malt agar freshly grown
fungal colony in the centre of each plate. The cultures were kept in a growth chamber at 22°C at
70% HR. Growth was evaluated every 2 or 3 days by measuring the diameter of the colony
estimated from the mean of two perpendicular diameters and expressed as a percentage of the
room available for growth, i.e. the diameter of the dish. Growth inhibition was calculated
according to the formula :
Growth Inhibition (%) = 100 x (1 – d1/d0)
where d0 is the diameter of the control culture and d1 the diameter of the culture in the presence
of extracts. Growth inhibition was calculated when the diameter of the control culture reached
9 cm. All experiments are repeated two times. Additional controls realized with small amount of
ethanol showed no significant difference with those containing only malt agar. Fungi utilised for
growth inhibition tests were Coriolus versicolor, Poria placenta and Aureobasidium pullulans. Four
types of extractives were systematically investigated resulting from soxhlet extraction with
dichloromethane, acetone, toluene/ethanol mixture (2/1) and water.
3.3.3. Measurement of efficacy against termites
A standard method for laboratory evaluation to determine resistance to subterranean termites
AWPA E1-97 was adapted. Heartwood test samples were extracted using dichloromethane,
acetone, water and mixture of toluene/ethanol. They were exposed in the laboratory to
Macrotermes natalensis found in East Africa. Pinus sylvestris was used as control.
The test was done in chambers free of organic material in glass jars of diameter 80 mm by
100 mm in height. Prior to use, all containers were sterilized in the autoclave. One hundred and fifty
30
grams of Fontainebleau sand were added to each container, followed by 30 ml of distilled water then
allowed to set for two hours.
Two test blocks 30 x 10 x 20 mm (l, r, t) were placed on sand in each jar with two corners
against the side of the container. Four hundred termites were counted (figure 10) and put to each
container at a ratio of 360:40 workers to soldiers respectively. All containers were incubated at
25°C for 28 days.
Figure 10. Collection and counting of termites
Classification of wood durability against termites was done according to EN 117 standard as
described in table 1.
Table 1. Classification of wood durability against termites
Block aspect after test Classification
No attack
Attempt of attack
Weakly attacked
Moderately attacked
Strongly attacked
0
1
2
3
4
31
Percentage of live termites and weight losses (WL) of the blocks were determined after the
experiment.
WL (%) = 100xm
)mm(
1
10 −
where m0, is the initial dried weight of the block and m1, the dried weight of the block after exposition
to the termites.
3.4. Determination of dimensional stability
Twenty four samples of Prunus africana were cut into regular blocks. All the surfaces
were smoothened using a planer. Dimensions (length, width and height) were measured to the
nearest 0.01 mm using Veneer callipers after drying to constant weight at 103°C and volume 1
determined (V1).
The blocks were then put in desiccators containing a saturated copper sulphate solution.
Weight of these blocks was measured every two days until stabilization to constant mass
indicating that the wood have attained their maximum moisture level.
Dimensions were measured again and volume 2 determined (V2). Swelling coefficient (S)
was determined according to the formula :
S (%) 100xV
VV
1
12 −=
3.5. Chemical composition
3.5.1. Determination of Klason lignin content (ASTM, 1998)
Before determination of lignin content, sawdust was prealably soxhlet extracted with
successively ethanol and toluene/ethanol mixture (1/2) for 8 hours respectively. Extracted sawdust
was then dried at 103°C. 500 mg of extractive free dry sawdust (m0) was introduced in 10 ml of 72%
concentrated sulphuric acid at room temperature and stirred for 4 hours. The mixture was then
32
diluted with 240 ml of distilled water and heated at 100°C for 4 hours by the mean of an oil bath.
The solution was left for 25 minutes to cool, then filtered through a Buchner furnel, rinsed with hot
water (70°C) and the residue dried 103°C to constant weight (ml).
Klason lignin content was calculated according to the formula :
KL (%) 100xm
m
0
l=
3.5.2. Determination of mineral content
1 g of dried sawdust (m0) was heated in a furnace at 500°C for 4 hours. The residue of ash
was weighed (ma) and the mineral content (ash %) determined according to the formula :
Ash (%) 100xm
m
0
a=
3.6. Scanning electron microscopy
Microscopic observations were performed with an environmental scanning electron
microscope (ESEM Quanta 200). The transverse surface of test sample was simply microtomed and
analyzed without further preparation. To be able to observe the same area before and after extraction
with dichloromethane transversal surface was clearly repeared with mark. Photomicrographs were
taken at different magnifications.
3.7. Spectroscopic analysis
3.7.1. NMR Analysis
1H NMR analysis was recorded in CDCl3 on a Bruker AM 400 spectrometer for Prunus
africana dichloromethane extract. 50 mg of this extract was dissolved in dimethylsulfoxide (DMSO-
d6) with tetramethylsilane (TMS) as internal reference. Chemical shifts were expressed in ppm and
calculated relative to TMS.
33
3.7.2. Infrared analysis
Infrared analyses were performed on Prunus africana dichloromethane extract using Perkin
Elmer FTIR spectrometer, SPECTRUM 2000. 9 mg of the extract were mixed homogeneously with
KBr (300 mg) and pressed at 450 bars for 5 minutes to form pellets. Spectra were recorded on a
wave range of 4000-400 cm-1.
34
4. Results and discussion
4.1. Soxhlet extraction
Table 2 shows the percentage of extractives obtained from Prunus africana heartwood and
sapwood by soxhlet extractor using different solvents. The measurement was done directly (DM)
and indirectly (IM) for all the samples.
Table 2. Percentage wood extracts from Prunus africana heartwood and sapwood by soxhlet
Solvent Heartwood Sapwood
Direct method Indirect method Direct method Indirect method
Hexane 0.3 0.4 0.3 0.5
Dichloromethane 1.1 1.4 1.1 1.6
Toluene/Ethanol 4.4 4.7 4.0 4.5
Acetone 3.4 3.9 3.0 3.6
Water 4.8 5.0 4.4 5.0
The results show that in all cases, indirect measurements (IM) gave higher extracts
content than direct measurements (DM). This may have been due to loss of some extractive
components during the removal of solvent in the Rotavapor or manipulation of the sample.
Table 3 shows the results of soxhlet extractions realized successively on the same batch of
sawdust with different solvents of increasing polarity used in the order listed in the table.
Table 3. Series extractions with different solvents of increasing polarity (Direct method)
Solvent Heartwood Sapwood
Hexane 0.3 0.3
Dichloromethane 0.4 0.4
Acetone 3.0 2.9
Toluene/Ethanol 3.4 3.0
Total Ext 7.2 6.5
35
The quantity of extractives increases with polarity of the solvent used and these results are
in agreement with observations performed on other tropical wood species. Hexane recorded the
lowest percentage of 0.3 to 0.4% extract followed by dichloromethane with 1.1 to 1.6%. Acetone,
toluene/ethanol and water had above 3% extracts content. Water extraction leads to the highest
extracts contents ranging values of 5%. Generally, heartwood leads to slightly higher percentage
of extractives compared to sapwood. High quantities of heartwood extracts may be, as described
in the literature on other tropical species, one of the reasons for wood durability (Neya et al.,
2004).
4.2. Biological tests
4.2.1. Natural durability against fungi
Results showed that Prunus africana has high natural durability against Coriolus
versicolor after three months exposure in Pétri dishes. Samples were observed to have minimal
surface attack as shown in figure 11.
Figure 11. Prunus africana wood blocks and beech controls after 3 months test against Coriolus
versicolor. a : undried Prunus africana blocks, b : 103°C ovendried Prunus africana blocks,
c : 103°C ovendried beech controls
a b c
36
In all cases, no significant weight losses were observed for Prunus africana heartwood
blocks after 16 weeks exposure to Coriolus versicolor (table 4). In the same time, beech controls
were strongly degraded indicating the virulence of the fungal strain used confirming the durability of
Prunus africana heartwood.
Table 4. Weight loss for Prunus africana after 16 weeks exposure to Coriolus versicolor
Sample Weight loss (%) Moisture content (%)
Undried Prunus africana blocks 0 41
103°C ovendried Prunus africana blocks 2.3 43
103°C ovendried beech blocks 27 72
The natural durability was determined based on theoretical weight to avoid modification
of extractives during the drying process. This test reflects the natural situation in the field. It was
observed that oven dried samples were slightly more attacked as compared to samples which
were not dried. This observation confirms that drying the wood at 103°C modified slightly
extractives composition reducing their effect against fungal attack.
Moisture content recorded at the end of the experiment is lower in Prunus africana blocks
compared to beech blocks, but sufficient to allow fungal development indicating that other
factors apart from moisture uptake are involved in durability.
4.2.2. Inhibition of fungal growth by extractives
To evaluate the effects of extractives on natural durability of Prunus africana heartwood,
fungal growth inhibition tests were investigated. Wood extractives have been reported to inhibit
fungal growth and hence contribute to natural durability of wood (Reyes Chilpa et al., 1998 ;
Celimene et al., 1999 ; Mori et al., 1997 ; Haupt et al., 2003 ; Windeisen et al., 2002 ; Neya et
al., 2002 ; Gérardin et al., 2004).
37
Growth inhibition values recorded after total colonization of the control Petri dishes (10
days for Coriolus versicolor, 14 days for Poria placenta and 16 days for Aureobasidium
pullulans) are presented in figure 12.
Figure 12. Effect of the different extracts tested at 100 and 500 ppm on mycelium growth
Growth inhibition depends of the solvent used for extraction. Independent of the fungus
tested, dichloromethane extracts presented higher activities. The inhibition was total for the three
fungi at concentration of 500 ppm. At lower extracts concentration (100 ppm), this latter one is
complete for the blue stain fungus Aureobasidium pullulans and the brown rot fungus Poria
placenta, while only partial inhibition was observed for the white rot fungus Coriolus versicolor.
Toluene/ethanol and acetone extracts present lower activities allowing partial inhibition of the
different fungi excepted in the case of Aureobasidium pullulans at 500 ppm. Water extracts
presented high activities at 500 ppm, while lower concentration lead to partial inhibition.
Behaviour of the different extracts tested at 100 ppm as a function of time is reported in figure
13.
0
10
20
30
40
50
60
70
80
90
100
Coriolus versicolor Poria placenta Aureobasidiumpullulans
Gro
wth
Inhi
bitio
n (%
) Dichloromethane 100 ppm
Dichloromethane 500 ppm
Acetone 100 ppm
Acetone 500 ppm
Toluene/ethanol 100 ppm
Toluene/ethanol 500 ppm
Water 100 ppm
Water 500 ppm
38
0102030405060708090
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Coriolus versicolor 100 ppm
dia
met
er (
mm
)
0102030405060708090
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Coriolus versicolor 500 ppm
dia
met
er (
mm
)
01020
30405060
708090
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Poria placenta 100 ppm
dia
met
er (
mm
)
0102030405060708090
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Poria placenta 500 ppm
dia
met
er (
mm
)
0
10
20
30
40
50
60
70
80
90
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Aureobasidium pullulans 100 ppm
dia
met
er (
mm
)
0102030405060708090
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Aureobasidium pullulans 500 ppm
dia
met
er (
mm
)
days days
days days
days days
Figure 13. Effect of extractives on growth of different fungi
Control Acetone Toluene/Ethanol Dichloromethane Water
39
Development of the mycelium on the medium treated by toluene/ethanol, acetone or water
extracts started after a more or less important inhibition period. During this period, activity of the
fungus was detected by the formation of a coloured area around the fungal inoculate (figure 14).
This behaviour is probably associated to a detoxification of the medium by fungal enzymes,
allowing further development of the fungus. According to these observations, it seems that
toluene/ethanol, acetone and water extracts possess only fungistatic properties. Dichloromethane
extracts exhibited higher activities leading to important inhibition of the fungal growth even at
low concentration.
Control 100 ppm 500 ppm
Fig. 14. Inhibition of Aureobasidium pullulans growth by different concentrations
of toluene/ethanol extracts after12 days
Generally, Aureobasidium pullulans (AP), showed the lowest resistance against extractives,
compared to Coriolus versicolor and Poria placenta. Sapwood extracts (data not shown) showed
lower growth inhibition than heartwood extracts. It was observed that Coriolus versicolor had
some limited growth in malt agar containing acetone and toluene/ethanol sapwood extracts at
500 ppm whereas at this concentration, heartwood extracts showed total inhibition. This may be
due to difference in extractive chemical composition between heartwood and sapwood.
4.2.3. Exposure to termites in laboratory
Behaviour of Prunus africana heartwood exposed to termites attack is reported in table 5.
40
Table 5. Resistance of Prunus africana wood samples
against Macrotermes natalensis termite species.
Samples Degree of
attack
Workers
(% Survivors)
Soldiers
(% Survivors)
WL (%) (1)
Water extracted 2 6 5 22.2 ± 0.6 Acetone extracted 1 4 5 13.6 ± 0.4 Dichloromethane extracted 2 3 7.5 23.6 ± 0.5 Toluene/ethanol extracted 1 0 0 10.3 ± 0.4 Un-extracted 0 0 0 0.3 ± 0.1 Control Pinus sylvestris 4 67 55 31.8 ± 2.2
(1) Average value of 4 replicates
Results showed that Prunus africana was naturally durable against termites. Termite
activity in the test bottles was noted to decrease with time. The highest mortality rate was
observed for un-extracted Prunus africana samples showing the importance of extractives on the
high mortality rate noted. All the termites in the un-extracted sample died after the first 15 days.
Weight losses (WL) measured after exposure of Prunus africana wood blocks and the controls to
termites in the laboratory is consistent with the degree of attack estimated visually and reported in
the same table. The percentage of survivor termites after the experimental period of 28 days also
reported in the table is also in agreement with observed weight losses.
Samples extracted by water or dichloromethane were more susceptible to termites attack
as demonstrated by the higher weight losses measured, compared to those extracted with acetone
or toluene/ethanol mixture. In the same time, un-extracted samples showed the highest resistance
against termite. Pine control had the lowest mortality rate and presents the highest degree of
attack. Similarly to the results obtained on the effects of extractives on fungal growth inhibition,
these results clearly demonstrate the importance of extractives on durability against termites.
41
4.3. Dimensional stability
Prunus africana’s swelling coefficient was equal to 4.5%. This relatively low value
indicates an important dimensional stability of Prunus africana heartwood, which could be
partially associated with wood durability. Indeed, wood dimensional stability is often linked to
lower water uptake as demonstrated in test against fungi. Even, if the moisture content of 41%
recorded previously (cf. table 4) could be considered as sufficient for the development of fungi, it
could be also in part involved, with the higher dimensional stability of wood, in an higher
difficulty for the fungi to colonize wood.
4.4. Chemical composition of Prunus africana
Chemical composition of Prunus africana heartwood is reported in table 6.
Table 6. Chemical composition of Prunus africana
Composition (%)
Lignin (%) 37.5
Polysaccharides (%) 54.5
Extractives (%) 7.2
Ash (%) 0.8
Each value in the table is average of 4 test samples.
Results reported in table 6 show that Prunus africana possess a relatively high lignin
content. Lignin has a complex, non-repetitive three-dimensional structure, which renders it
resistant to attack by numerous micro-organisms. The only organisms capable of mineralizing
lignin into water and carbon dioxide are a select group of basidiomycetes, the so-called white-rot
fungi. However, even white-rots cannot live on lignin alone but require other, more easily utilized
carbon sources to sustain lignin degradation which is thus said to be co-metabolic (Anke et al.,
2006). Therefore high contents of lignin would partly explain high durability of Prunus africana.
High content of extractives, which is usual for tropical species, could also be involve in the
reasons of the higher natural durability of Prunus africana.
42
4.5. Scanning electron microscopy
ESEM observation of transversal section of Prunus africana heartwood is reported
figure 15.
A B
C D
Figure 15. ESEM analysis of transversal section of Prunus africana
A. Vessels mainly in radial arrangement. B. Vessels magnification showing extractives
B. deposits in the lumina. C. Ray magnification showing extractives deposits. D. Fibers
43
Wood presents a porous structure without growth rings. Vessels are disposed mainly in
radial arrangement solitary, in pairs, in groups of 3 to 5, and clusters of up to 5. The diameter of
vessels is comprised between 50 and 150 µm. The fibers are very abundant and thick-walled.
Their thickness is comprised between 10 and 30 µm. Parenchyma is present in lower proportion.
Deposits of extractives are abundantly present in heartwood vessels and also obvious in ray cells.
Observations performed in tangential section are reported in figure 16.
A B
Figure 16. ESEM analysis of tangential section of Prunus africana
A. Vessels, rays and fibers. B. Ray cross section showing extractives deposits
Analysis of tangential section confirms the presence of extractives in rays. Rays are
generally multiseriate , 3 to 6 cells wide and up to 15 cells high.
To establish the presence of extractives in the vessels, microscopic observations have
been performed after dichloromethane extraction of the small piece of wood used for anatomical
characterization described above. Results obtained are presented in figure 17
44
.
Figure 17. ESEM analysis of transversal section of Prunus africana
after dichloromethane extraction
Observations performed on the same vessels (compare figure 15A, B and figure 17)
indicate that most of the deposits present in the vessels before extraction disappear after soxhlet
extraction.
Since dichloromethane extract showed higher inhibition rate for fungi as compared with
acetone, toluene/ethanol and water extracts, the extractives deposited in the cells play probably a
vital role in the durability of Prunus africana wood species.
4.6. Spectra analysis
4.6.1. 1H NMR analysis
To identify the nature of extractives responsible of durability, spectroscopic analysis was
performed on the more active dichloromethane extracts. Figure 18 present 1H NMR spectrum of
the light brown solid obtained after extraction and evaporation of dichloromethane under
vacuum.
45
1 .02 .03 .04 .05 .06 .07 .08 .09 .01 0 .01 1 .01 2 .0
0.94
1.04
1.08
4.28
47.72
13.18
0.37
D M SO
Water
3 .5 04 .004 .5 05 .0 0
0.80
0.65
0.55
0.43
1.38
0.57
0.51
0.57
0.13
Figure 18. 1H NMR spectra of Prunus africana dichloromethane extract
On the basis of 1H NMR spectrum and of literature data reported on chemical composition
of bark extractives (Catalano et al., 1984 ; Fourneau et al., 1996), it seems reasonable to assume
that the main compounds present in dichloromethane extract are constituted of phytosterols.
Signal at 12 ppm corresponds to the hydrogen of a carboxylic acid function. The singlet at
5.25 ppm is ascribable to a vinylic proton present on a tri-substituted double bond. Signals
comprised between 4.3 and 3.6 ppm are attributed to the presence of a sugar unit, while those
comprised between 2.1 and 0.8 ppm are characteristic of an aliphatic structure. According to
these observations and to results reported in literature on the chemical composition of bark, it is
assumed that dichloromethane extracts consisted mainly of glycoside of triterpenic acids like
oleanolic and ursolic acid.
46
COOH
Sugar-O
COOH
Sugar-O
oleanolic acid glycoside ursolic acid glycoside
Figure 19. Possible structure of main components of dichloromethane
heartwood extracts of Prunus africana
4.6.2. Infrared analysis
Figure 20. FTIR spectrum of dichloromethane extractives of Prunus Africana
FTIR spectrum for Prunus africana dichloromethane is presented in Figure 20. It
confirms previous NMR observations: signals at 1690 cm-1 and 2600 cm
-1 (broad absorption) are
characteristic of carbonyl and hydroxyl groups respectively of carboxylic acid function. Hydroxyl
groups of sugar unit appear at 3400 cm-1, while strong absorption at 2860 cm
-1 is characteristic of
C-H vibrations present in aliphatic structure of terpenes. Identification of oleanolic and ursolic
4000,0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 500,0
18,7
30
40
50
60
69,6
cm-1
%T %T
1690
3400
2600
2860
47
acid is consistent with literature data concerning antifungal activities of these products (Becker et
al., 2005 ; Deepak and Handa, 2000). Moreover, saponins derived from oleanolic acid are also
described to possess antifungal properties against phytopathogenic fungi allowing to explain the
durability of Prunus africana heartwood to fungal degradation (Escalente et al., 2002)
48
5. Conclusion
Results showed that Prunus africana heartwood is very resistant to termites and fungal attacks.
Wood durability against fungi seems in great part due to the presence of extractives, which have been
shown to inhibit the growth of several fungal species among which Coriolus versicolor and Poria
placenta, two important rotting fungi. Dichloromethane and water extractives present the higher
fungicidal activity allowing total growth inhibition of all tested species at concentration of 500 ppm,
while acetone and toluene/ethanol extractives lead only to partial inhibition. Durability of Prunus
africana heartwood to termites is correlated with the presence of extractives. Unextracted heartwood
present an important durability, while extracted samples present lower durability. Higher susceptibility
to termites is observed for water and dichloromethane extracted heartwood indicating the importance
of the extractives present in these fractions on natural durability. 1HNMR and FTIR analysis of the
more active dichloromethane extracts indicates the presence of phytosterol glycosides. Comparison
with literature data reported on the chemical composition of Prunus Africana bark allows to identify
triterpenic acids like oleanolic and ursolic acid as the main components. Moreover, Prunus africana
heartwood possess high lignin content, high dimensional stability and lower water uptake after fungal
exposure, which could be also involved in the reason of natural stability.
Recommendations and prospects
1) Chemical analysis of Prunus africana extractive should be continued by chromatographic
methods like HPLC-MS or GC-MS to identify exact composition.
2) Termiticidal properties of dichloromethane extracts should be investigated and compared to
commercially available products to evaluate the potentiality of such compounds.
3) Durability test of Prunus africana against other biodegraders like bacteria and insects should be
tested.
4) Valorization of other durable wood species from Tropical countries for example, Meru oak,
Elgon teak and African mahogany.
49
Part 2. Improvement of Grevillea robusta durabilily
Introduction
About Grevillea robusta, justification of its treatment
The second part of this work involved treatment of Grevillea robusta species to improve its
durability. Grevillea robusta species is an agro-forest tree grown in the Central, Eastern, Nyanza
and Rift-Valley provinces of Kenya (figure 21). It is a fast growing hardwood, which is not
durable against biodegraders (Okario and Peden,1992). Most of the fast growing trees contain
high ratio of sapwood to heartwood hence lower natural durability than selected slow-grown trees
extracted from natural forests (Kamweti, 1992). Generally sapwood is more vulnerable to
biological degradation than heartwood for most wood species (Grace, 1996). Both heartwood and
sapwood of Grevillea robusta species are susceptible to fungi, termite and powder-post-beetle
infestation (Kamweti, 1992 ; Floyd, 1989).
In Kenya, forest resources rank high among other important natural resources. Forest
ecosystems are complex natural resource base which provides environmental goods and services
for social, cultural and economic development. It is therefore important that the resource is
conserved, protected and sustainably utilized for national development. Kenya like other
developing countries is confronted with challenges of over reliance on natural resources. Timber
demand in Kenya is rising each year unlike its supply. If this trend continues, it is projected that
by the year 2020, Kenya will have a deficit of 6,841,000 m3 of wood (KFMP, 1994). The deficit
is increased by premature failure of wood in service as a result of biological deterioration,
failures in establishing new and expanding forest plantations and changes of forest land use. One
way to solve this problem of timber deficit and deforestation is to increase wood durability in
service by effective treatment.
50
Figure 21. Distribution of Grevillea Robusta in Kenya
NAIROBI
ETHIOPIA
MANDERA
MARSABI
WAJIR
TURKANA
ISIOLO SAMBURU
GARISA
TANA RIVER LAMU
KILIFI
KWALE
TAITA TAVETA
KAJIADO MAKUENI
MACHAKOS
KIAMBU MURANGA
KITUI
EMBU
THARAK A NITHI
MERU
KIRINYAGA NYERI
NY
AN
DA
RU
A
NAKURU
NAROK
MIGORI
HOMA BAY
KISUMU
KISII
KERICH
NAND
BOME
SIAYA
KAKAMEGA
UASIN GISHU
VIHIGA
NY
AM
IR
BARING
LAIKIPIA
BUNGOMA
BUSIA
TRANS
WEST POKOT
S U D A N
UGANDA
TANZANIA
Indian Ocean
MOMBASA
KEY
S O
M A
L I
ELG
EY
O M
AR
AK
WE
T
Grevillea robusta
AFRICA
International boundary
District boundary
51
The Kenyan, forest plantations consists of 31% Pines, 45% Cypress, 10% Eucalypts and
others 14% being the main base for industrial raw materials before logging ban was effected by
the government in 1999. Private forests cover about 70,000 ha consisting mainly acacia species
and Grevillea robusta out of which the later covers about 75% of the area (KFMP, 1994). Acacia
is used for tannin extraction while Grevillea robusta (figure 22) is multipurpose agro-forest tree.
Use of Grevillea robusta for fencing and construction purposes has been necessitated by wood
deficit in Kenya that started from the year 2000 due to industrial development and logging ban
from government forests. Increase of population in Kenya has also caused depletion of natural
forests leading to shortage in the industrial raw material. This situation has forced farmers to
deviate from tradition and utilize other wood species for fencing and construction purposes. Most
of them have resulted to use Grevillea robusta due to availability but its low durability is a great
concern to the government.
Figure 22. Grevillea robusta trees, 20 years old on agricultural land in Central province, Kenya.
52
Sap-displacement treatment method uses the principle of hydrostatic pressure to force the
preservative from the butt end of the round timber to the top. Some impermeable species like
Picea species take long time to season adequately. Fowlie and Sheard (1983) reported successful
treatment of Picea abies, Picea sitchensis and Pinus nigra using pressurized sap displacement in
20-30 hours. This method gives a greater concentration of the preservative at the butt end and this
is added advantage especially in the case of fencing poles used in ground contact and most
vulnerable to decay. It is also economically viable in remote areas where impregnation facilities
are lacking and the cost of labour is low. Effectiveness of these treatments depends not only on
the method but also of the preservative employed. It is generally accepted that sap displacement
allows treating sapwood effectively allowing homogeneous penetration of the active ingredients
used for wood preservation. Due to the importance of decay and termites degradation, CCA was
chosen as the more effective relatively low cost preservative to improve durability and
performance of Grevillea robusta as a locally available construction material.
There have been no attempts yet of looking into appropriate cheaper chemicals and
simpler techniques to effectively treat poles and posts from the available species. It is also true
that pressure treatment of fencing posts is expensive and beyond the financial abilities of most
local people. The economics and feasibility of non-pressure on-site treatments using available
preservatives have never been investigated. In this context sap displacement method using CCA
was tested on Grevillea robusta round wood samples as a cheap method of treatment applicable
on site. Results of CCA retention and penetration were compared with the FAO
recommendations for different hazard areas. Effective treatment could reduce replacement costs
and provide durable construction material in Kenyan rural areas where Grevillea robusta is in
high demand for construction timber and fencing purposes. This would play a very important role
in forest conservation.
Heat treatment was also performed on Grevillea robusta as more environmentally
acceptable modern technique of wood preservation. The concern about environmental
degradation has increased in the recent past. This has led to an important change in the field of
wood preservation in regard to biocide toxicity leading to development of non-biocidal
alternatives mainly heat treatment of wood by mild pyrolysis in Europe. This type of wood
53
treatment has not been investigated on Kenyan wood species. However, it is important to keep in
mind that such treatments could be of valuable interest to increase durability of non durable wood
species for applications in hazard class 3 ; but probably not for hazard class 4, where wood in soil
contact, is subjected to severe degradation conditions.
54
Part 2. Chapter 1. Impregnability of Grevillea robusta
using sap displacement method
1. Introduction
Wood is widely used for construction due to its physical characteristics and engineering
properties associated with its high availability (Eaton and Hale, 1993 ; Aston, 1985). As the
technology advances, more diverse and complex uses of wood have been developed. Despite its
versatility as a construction material, modern technology has brought other alternatives that have
superseded wood in some qualities like metal and plastics.
The natural durability of timber is fairly often a disadvantage of wood compared to other
concurrent construction materials. Durability is mainly dependent on the wood structure and
chemical composition of extractives of individual species (Fengel and Wegener, 1984). The
heartwood was generally described to be more resistant than sapwood to fungi and insect attacks.
Use of non naturally durable species like Grevillea for fencing and construction purposes
has been necessitated by wood deficit in Kenya that started from the year 2000 due to industrial
development and logging ban from government forests since the year 2001. The increase in
population has caused depletion of the natural forests leading to shortage in industrial raw
material. As a result of this and the current ban on logging in Kenya, consumers have sought
alternative sources and species for manufacturing and constuction including use of Grevillea
robusta. Grevillea robusta is a fast growing agro-forestry tree species distributed in most parts of
Central and Rift Valley provinces in Kenya (KFMP, 1994). Most of the fast growing trees
contain high ratio of sapwood to heartwood hence lower natural durability than selected slow-
grown trees extracted from natural forests.
Presently, three treating chemicals namely copper chrome arsenate (CCA),
pentachlorophenol (PCP) and creosote oil are used in Kenya (Venkatasamy, 1997 and 2000).
Treatment techniques include pressure, hot and cold process and dipping process. Cheap and
55
simple techniques of treatment have not been introduced in the country especially in the rural
areas, where about 80% of the population is located.
Sap-displacement method was applied on freshly felled Grevillea robusta using the
principle of hydrostatic pressure to force the preservative from the butt end of the round timber to
the top. Due to the importance of decay and termites degradations, CCA was retained as the more
effective low cost preservative to improve durability and performance of Grevillea robusta as a
construction material.
The aim of this study was to investigate the impregnability of Grevillea robusta using
sap-displacement to enhance its uses, reduce replacement costs and provide a relatively durable
construction material especially in rural areas where it is mostly grown as an agro-forestry tree.
56
2. Materials and methods
2.1. Sampling
Specimens from twenty years old freshly cut round wood of Grevillea robusta A. Cunn
ex. R. Br. with a top diameter of 140-180 mm and 1m length were cut following the procedure of
Harwood and Booth (1992). Sixty specimens were prepared: thirty debarked and thirty
undebarked but with a small debarked portion to allow preservative uptake. For barked wood, the
debarked portion is dipped in the container with CCA to avoid direct absorption of the chemical
by the bark. The uptake of CCA from the container was expected to be mainly by suction created
through sap displacement.
2.2. Determination of fungal susceptibility of Grevillea robusta
Grevillea robusta samples were prepared from twenty four-year-old trees, chosen at
random from Central, Rift valley and Eastern provinces in Kenya. These were cut into 25 x 25 x
50 mm blocks. They were dried to constant moisture content in the oven at 60°C. The standard
method AWPA E14-94 was adapted for this experiment. Moist forest soil was put in six plastic
buckets (figure 23) and left to condition for 4 hours. Twelve sample blocks were buried in each
bucket and kept at room temperature in the laboratory. Control samples were buried in sterilized
sand, free of organic matter.
Weight loss (WL) was determined after every one month by removing 12 sample blocks
using the following formula
100xm
)mm((%) WL
0
10 −=
where m0 is the oven dry mass before attack
m1 is the oven dry mass of block after attack.
57
Figure 23. Forest soil containing Grevillea robusta test samples against fungi
2.3. Sap-displacement treatment
Industrial quality of Chromated Copper Arsenate (CCA) was used during this study
(Tanalith-C, Arch Timber). Aqueous solutions of CCA were prepared at different concentrations
(2, 4 or 6% in mass) and used to impregnate the specimens. The butt end of each specimen was
placed in plastic container filled with CCA solution (5 l for debarked and 4 l for undebarked
poles). The container was properly tied on the butt end and supported in a slanting position as
shown in figure 24.
The treatment set-up was maintained until chemical oozed out from the top end of the
posts or when all the preservative in the container was absorbed. At the end of the treatment the
containers were removed and specimens left in the same position for 7 days to leach-out unfixed
preservative. Retention (kg/m3) was determined for both debarked and un-debarked specimens
and compared with the recommended average retention for timber under various exposure
conditions according to the formula :
Retention = (m2-m1) × c / v
58
where m1 is the mass of the fresh untreated sample (kg), m2 is the weight of impregnated sample
measured immediately after the CCA solution had oozed out from the top of the pole (kg), c is
the CCA solution concentration (%) and v the sample volume (m3)
Penetration was measured in percentage along the height of samples according to the
formula :
Penetration = (1-(r-r1)/r ) × 100
where r1 is the depth of preservative penetration (cm) and r the radius of the sample (cm) for a
given position from butt end.
Figure 24. Treatment of undebarked specimens
59
3. Results and discussion
3.1. Fungal susceptibility
Grevillea robusta was noted to be highly susceptible to fungal attack. Dark brown
colouration, a sign fungal colonization was observed after three weeks in the forest soil. Colouration
increased with exposure time as shown in Figure 25.
Figure 25. Grevillea robusta samples showing variation in colour due to fungal colonization
after 6 weeks and 3 weeks compared with control.
Three types of fungal decay, namely, soft rot, brown rot and white rot are known to be
dominant in forest soils in Kenya (KFMP, 1994).
Weight loss increased with time as shown in figure 26. This could be associated to increase
in colonization with time as evidenced by colour change.
60
0
5
10
15
20
25
30
WL
(%
)
Figure 26. Variation of Grevillea robusta weight loss (%) with time, in the forest soils
3.2. Preservative retention
According to Food and Agriculture Organisation (FAO), the recommended retention of
CCA for interior timber not in ground contact such as trusses, rafters is 6 kg/m3 and exterior
timber not in ground contact for example doors and windows is 8 kg/m3. Timber in ground
contact such as fence posts, railway sleepers, bridge timbers is 12 kg/m3. Timber permanently
immersed in fresh waters require retentions of 16 kg/m3 and timber immersed in seawater e.g.
groynes, jetties, boat building timber require retentions of 24 kg/m3 (FAO, 1986).
Table 7 reports mean CCA retention values obtained for debarked and undebarked
specimens and their possible utilizations according to FAO recommendations.
Debarked specimens had significantly higher retention than the un-debarked specimens
for the same solution strength. In both debarked and undebarked specimens, retention increases
with the strength of the solution.
1 2 3 4 5 t (months)
61
Table 7. Mean CCA Retention values measured and possible uses according to FAO standards
Debarked specimens Undebarked specimens
Treatment
solution
concentration
(%)
CCA Retention
(kg/m3)
Possible use
according to FAO
CCA Retention
(kg/m3)
Possible use
according to FAO
2 16.66 + 1.70 Permanently
immersed in fresh
water
12.55 + 1.73 Timber in contact
with the ground
4 32.69 + 2.97* Permanently
immersed in sea
water
16.60 + 1.68 Permanently
immersed in fresh
water
6 37.66 + 5.60* Permanently
immersed in sea
water
25.04 + 3.26 Permanently
immersed in sea
water
Key: * indicates significantly different values at 5% level of significance
Comparisons between the experimental mean retention values and the FAO recommended
values indicated that impregnabibility of G. robusta using sap displacement method is sufficient
to reach the value of 12 kg/m3 necessary for uses in ground contact. For the 2% solution strength,
debarked specimens did not have a significantly different value from the value recommended for
immersion in fresh water. This implies that G. robusta meets the minimum requirements set for
wood to be used permanently submerged in fresh water. However, the values obtained for the 4%
and 6 % concentration were significantly higher than the recommended values. These are suitable
for all environmental hazards but they are uneconomical because of high retention obtained.
Values obtained for un-debarked specimens using the 2, 4 and 6% solution strengths were
generally smaller than those obtained for debarked specimens. This may be due to the increased
surface area hence more suction pressure which lead to chemical absorption. However, these
values are in all cases higher than the 12 kg/m3 recommended by FAO. G. robusta treated at 2%
and 4 % solutions concentrations can meet the minimum retention requirements for utilization in
ground contact and submersion in fresh water respectively. Values obtained for the 6 % CCA
solution strength were slightly higher and allow utilization of wood under permanent submersion
in marine water. Figure 27 shows the relationship between retention and solution strength for
debarked and un-debarked samples.
62
y = 3.1225x + 5.5733R2 = 0.9604
y = 5.25x + 8.0033R2 = 0.9154
0
5
10
15
20
25
30
35
40
45
0 1 2 3 4 5 6 7CCA solution concentration (%)
Ret
entio
n (K
g/m
3)
Figure 27. Relationships between retention and CCA solution concentration
The linear relationship is described by equations (i) and (ii) and with the corresponding
correlation coefficients of R2 = 0.92 and R
2 = 0.96 respectively :
Retention = 5.25(Solution Strength) + 8.00 . . . . . . . . . . . . (i)
Retention = 3.12(Solution Strength) + 5.57 . . . . . . . . . . . . (ii)
The retention of un-debarked specimens has a stronger linear relationship than the
debarked specimens. These models can be used to determine the level of retention on both
debarked and un-debarked specimens using the sap-displacement method.
The higher retention values measured in debarked samples may be due to increased
surface area hence increased suction pressures which lead to evaporation leading to higher
chemical uptake.
3.3. Preservative Penetration
Table 8 shows the penetration for debarked and un-debarked specimens at 2%, 4% and
6% solution concentration.
63
Table 8. Mean penetration for debarked and undebarked specimens
Penetration CCA solution
concentration (%) Debarked Specimens Undebarked Specimens
2 80.6 + 12.7 63.7 + 21.7
4 85.6 + 9.8 71.1 + 17.6
6 89.4 + 7.8 73.2 + 6.9
For the three concentrations, debarked specimens had a higher penetration than un-
debarked ones.
Figure 28 shows the variation in preservative penetration with solution strength for
debarked and un-debarked samples.
Undebarked specimensy = 2.3657x + 59.837
R2 = 0.9047
Debarked specimensy = 2.2045x + 76.327
R2 = 0.9963
55
60
65
70
75
80
85
90
95
0 1 2 3 4 5 6 7
CCA solution concentration (%)
Pen
trat
ion
(%)
Figure 28. Relationships between depth of penetration and CCA concentration
measured at 500 mm height
The linear relationship is described by equations (i) and (ii) and with the corresponding
correlation coefficients of R2 = 0.97 and R
2 = 0.90 respectively :
Penetration = 2.21(Solution Strength) + 76.3 . . . . . . . . . . . . (i)
Penetration = 2.37(Solution Strength) + 59.8 . . . . . . . . . . . . (ii)
64
There were significant differences in penetration between debarked and un-debarked
specimens. Preservative penetration increased with an increase in solution strength. Debarked
specimens showed higher penetration than the un-debarked specimens. Debarked samples have
higher surface area for evaporation hence creating high suction pressure than barked samples
hence high penetration. Even in treatment of the same solution strength, samples showed
variation in penetration. This can be explained by difference in sapwood content, density and
moisture content. There is a stronger linear relationship between penetration and solution strength
for debarked than for un-debarked specimens. The strong correlation between penetration and
solution strength for the two groups of specimens indicate that the two models predict the degree
of preservative penetration adequately.
Variation of the depth of penetration of the preservative with the height of the specimen is
shown in figures 29 and 30 for debarked and undebarked samples respectively. In both cases,
penetration decreased with the height of the specimen. The depth of penetration was also
dependent of the treatment solution concentration, i.e. the higher is the concentration the higher is
the penetration and consequently the retention.
50
60
70
80
90
100
0 500 1000
height (mm)
Pen
etra
tion
(%)
treated at 2%
treated at 4%
treated at 6%
Figure 29. CCA penetration in function of extraction height for debarked specimens
65
30
40
50
60
70
80
90
100
0 200 400 600 800 1000
Height (mm)
Pen
etra
tion
(%)
treated at 2%treated at 4%treated at 6%
Figure 30. CCA penetration in function of extraction height for undebarked specimens
66
4. Conclusion
The results show that Grevillea robusta can adequately be treated to retention levels that
meet minimum requirements for various use conditions including ground contact or permanent
submersion in fresh and sea water. The positive linear relationship between retention, penetration
and solution strength means that the desired retention and penetration of preservative can be
achieved by varying the solution strength. Preservative penetration is dependent of the height of
the pole and inversely proportional to the distance from the end dipped in preservative.
Undebarked specimens tend to retain less preservative, while the debarked ones yield greater
penetration.
Debarked G. robusta poles can achieve retention required for wood permanently
submerged in fresh water at 2 % solution strength and for marine water exposure at solution
strength of 4% and 6%. The undebarked wood can be treated to retention levels required for
ground contact using 2% solution strength and for fresh water and marine exposure using 4% and
6% solution concentrations.
67
Part 2. Chapter 2. Evaluation of thermally modified Grevillea
robusta heartwood as an alternative to shortage of wood resource in
Kenya : characterisation of physicochemical properties and
improvement of bio-resistance
1. Introduction
The increasing environmental pressure and awareness over the last few years in Europe and
United states has led to an important change in the field of wood preservation in regard to biocide
toxicity (Preston, 2000). The pressure has resulted in voluntary removal of Copper Chrome
Arsenate (CCA) from wood preservatives in residential applications in the United States. A new
generation of copper organic preservatives was formulated as replacements, but these
preservatives may not provide a permanent solution to all related problems (Arango et al., 2006).
Such concern has lead to development of "non biocidal" alternatives like chemical or thermal
treatments (Hill, 2005). Heat treatment has been particularly developed in Europe during this last
decade leading to industrialisation and commercialization of heat treated timbers resulting from the
treatment of low natural durability wood species like pine, spruce, poplar or beech (Patzelt et al.,
2002 ; Alen et al., 2002)..
This form of treatment coverts wood into a new product called torrified or retified wood. The
change is through mild pyrolysis in a temperature range between 200 and 260°C under inert
atmosphere of Nitrogen. Such heat-treated wood dramatically reduces its hygroscopicity and
improves its dimensional stability and durability. The heat-treated wood becomes hydrophobic,
(Weiland and Guyonnet, 2003). The improved durability of heat-treated wood can be explained
in four ways according to Hakkou et al. (2005) :
- Increase of the hydrophobic character of wood, which limit the sorption of water into the material
hence not favourable for fungal growth.
- Generation of new extractives during heat treatment that act as fungicides.
68
- Modification of wood polymers leading to non-recognition of the latter by enzymes involved in
fungal degradation.
- Significant degradation of the hemicelluloses, which constitute one of the main nutritive sources
for fungi.
The retification process is mainly driven by transport phenomena (heat and mass transfer) and
pyrolysis kinetics. Parameters that influence this process include :
- Wood characteristics since wood components differ from hardwoods to softwoods, within species
and within geographical regions. Consequently wood properties greatly vary leading to different
retification schedules. Also the initial moisture content and its homogeneity within the load affect
total duration and quality.
- Geometrical parameters are very important for thickness above 1 mm.
- Chemical reactions in the reactor as a consequence of interaction of hot pyrolysis vapours with
decomposing wood solid. Wood dimensions, number of boards and pile stacking affect forced
convention and homogeneity of treatment within kiln.
- Temperature and duration of pyrolysis determine the quality of rectified wood.
Heat treatment in an oil bath is also a promising approach to upgrade wood for out-door
end use (Sailer et al., 2000). The treatment releases the stress in wood after removal of
hemicelluloses and degradation of lignin with the resultant wood being more resistant against
fungi (Kamdem et al., 2002).
It was important to investigate effect of heat treatment on durability of Grevillea robusta,
to allow its development and uses in Kenya for applications where mechanical properties are
secondary. Indeed, it is known that an important drawback of heat-treated wood is mechanical
brittleness (Santos, 2000 ; Mouras et al., 2002 ; Petrissans et al., 2003 ; Unsal and Ayrilmis,
69
2005 ; Yildiz et al., 2006). For these reasons, heat treated wood is not recommended for use in
load-bearing constructions, but can find valuable applications for furniture, wall, ceiling, roofing,
flooring and fencing. The extent of the change in timber properties during the heat treatment
depends on the method of thermal modification, the wood species and its characteristic
properties, the initial moisture content of wood, the surrounding atmosphere, treatment time and
temperature. Temperatures above 150°C alter the physical and chemical properties of wood
permanently and the mechanical properties start to weaken (Mitchell, 1988). The wood becomes
more brittle and bending and tension strength decreased but the dimensional stability increase
(Yildiz et al., 2006). It was therefore important to test the effects of heat-treated Grevillea
robusta on the modulus of rupture (MOR) and modulus of elasticity (MOE).
70
2. Materials and methods
2.1. Materials
Grevillea robusta heartwood was used in this study. Wood blocks measuring were cut and
oven dried at 103°C (approximately 48 hours) before determination of their anhydrous weights.
2.2. Heat treatment
Heat-treatment was performed on previously dried blocks (20 mm x 20 mm x 40 mm in
longitudinal, radial and tangential directions) in a reactor placed in an oven at different
temperatures (220, 240, 250 and 260°C) during different time (0.5, 1, 5, 7 and 15 hours) under a
nitrogen atmosphere. The oven temperature was increased by 20°C mn-1 from ambient to the
operating temperature. After the treatment, the temperature decreases slowly to the room
temperature. Mass loss after heat treatment was calculated according to the formula :
ML (%) = 100 x (m0 – m1)/ m0
where m0 is the initial oven dried mass of wood sample and m1 the oven dried mass of the same
sample after heat-treatment. Description of the equipment used for heat treatment is despicted in
figure 31.
Figure 31. Equipment used for heat treatment
TI
TI
FI
reactor
Wood sample
N 2
71
2.3. Exposure to fungi
Two kinds of experiments were performed to evaluate fungal durability after heat
treatment. The first was resistance evaluation of samples treated at 250°C for different duration
of time against Coriolus versicolor. The second was evaluation of heat treated blocks at 250°C
for 7 hours against different wood rotting fungal species, two white rots (Coriolus versicolor and
Pycnoporus sanguineus) and two brown rots (Poria placenta and Antrodia sp.). The choice of the
heat treatment conditions (250°C for 7 hours) was based on previously reported experiments
performed on beech indicating that improvement of durability can be achieved for mass losses of
approximately 20%. (Hakkou et al. 2006). Heat treated Grevillea robusta blocks were cut into
smaller blocks of 25 mm x 25 mm x 5mm in longitudinal, radial and tangential directions. These
were used for fungal durability evaluation after conditioning in an oven at 103°C for one night
(m2). Petri dishes (9cm diameter) were filled with sterile culture medium prepared by mixing
30 g malt and 40 g agar in distilled water (1l), inoculated with the different fungi and incubated at
22°C and 70% relative humidity to allow full colonization by the mycelium. Two blocks (treated
or untreated as control) were placed in each Petri dishes and incubated during 3 months to
evaluate the effect of heat treatment. Each experiment was duplicated or triplicated (see after).
After this period, mycelia were removed and the blocks were dried at 103°C and weighed (m3) to
determine the weight loss caused by the fungal attack.
WL (%) = 100 x (m2 – m3)/ m2
where m2 is the initial oven dried mass of wood block before attack and m3 is the oven dried
mass after attack.
Moisture content of the wood after fungal attack was determined as followed :
MC (%) = 100 x (m3’ – m3)/ m3
where m3’ is the wet weight of the sample after attack measured after removal of the mycelium
and m3 is the oven dried mass after attack.
72
2.4. Exposure to termites
Heat treated Grevillea robusta samples (7 hours at 250°C) were also exposed to
Macrotermes natalensis widely distributed in Kenya in laboratory conditions according to a
procedure adapted from AWPA E1-97 standard (Standard method, 1997). Pinus sylvestris and
untreated Grevillea robusta were used as control. The tests were done in chambers free of
organic material in glass jars 80 mm diameter by 100 mm in height. Prior to use, all containers
were sterilized in autoclave. 150 g of sand were added to each container, followed by 30 ml of
distilled water then allowed to stand for two hours. Two dried weighed (m2) test blocks, 30 mm x
10 mm x 20 mm in longitudinal, radial and tangential directions, were placed on the surface of each
jar with two corners against the side of the container. 400 termites were added to each container at a
ratio of 360:40 workers to soldiers respectively. All containers were maintained at 25° C for 28 days.
Percentage change in dry mass of the test block was determined. Blocks were then dried at 103°C
and weighed (m4) to determine the weight loss caused by the termite attack.
WL (%) = 100 x (m2 – m4)/ m2
where m2 is the initial oven dried mass of wood block before attack and m4 is the oven dried
mass after attack.
Resistance to termite was also investigated using field test conditions. Tests were carried
out according to a method adapted from AWPA E7-93 standard (Standard method, 1993)
especially as concerns the sample size which was reduced to fit with our heat treatment
equipment. An active termite nest constituted of Macrotermes natalensis was identified and
cleaned from all the cellulolytic materials. The site was then watered and covered with an opaque
plastic sheet to activate the termites. Samples measuring 50 mm x 25 mm x 25 mm in
longitudinal, radial and tangential directions respectively were buried down at a radius of 1m
from the termite nest. Three kinds of samples were used: 24 samples of heat treated Grevillea
robusta (7 hours at 250°C), 24 samples of untreated Grevillea robusta and 24 samples of Pinus
sylvestris. The moisture at the test site was improved by watering to maintain and attract more
termites. Evaluation was done every month by removing 6 stakes for each treatment and the values
averaged.
73
Classification of wood durability against termites was done according to EN 117 standard
Block aspect after test Classification
No attack
Attempt of attack
Weakly attacked
Moderately attacked
Strongly attacked
0
1
2
3
4
Weight loss (WL) of blocks due to termites attack was determined as follows
2.5. Determination of the amount of extractive
Untreated and heat treated Grevillea robusta blocks were ground to sawdust and dried at
103°C before extraction with dichloromethane in a soxhlet extractor for 16 hours at the rate of 8-
12 cycles per hour. After extraction the solvent was evaporated under reduced pressure and the
residue dried over P2O5 in a dessicator before weighing.
Direct and indirect measurements were made to determine the amount of extracts. The
methodology used is similar to that described before (cf. 3.2.2., part 1).
2.6. Spectroscopic analysis
2.6.1. FTIR analysis
FTIR spectra were recorded as KBr disks on a Perkin Elmer FTIR spectrometer
SPECTRUM 2000 between wave number range of 4000-400 cm-1. Finely divided 9 mg wood
samples, obtained after grinding of blocks and drying at 103°C, were dispersed in a matrix of
KBr (300 mg) and pressed to form pellets. Due to the difficulty to define a reference spectral
band, which remains completely invariable, to perform quantitative measurements, all
comparisons were made qualitatively.
74
2.6.2. CP/MAS 13C NMR analysis
Solid state CP/MAS (cross-polarisation/magic angle spinning) 13C NMR spectra were
recorded on a Bruker MSL 300 spectrometer at a frequency of 75.47 MHz. Acquisition time is of
0.026 s with number of transients of about 1200. All the spectra were run with a relaxation delay
of 5 s, CP time of 1 ms and spectral width of 20 000 Hz. Spinning rates are 5 KHz. Chemical
shifts are expressed in parts per million (ppm).
2.6.3. 1H NMR analysis
1H NMR spectra were recorded in CDCl3 on a Bruker AM 400 spectrometer. Chemical
shifts were expressed in ppm and calculated relative to TMS.
2.7. Contact angles measurements
Contact angles of untreated and heat treated blocks were recorded with a goniometer
using water as probe liquid. 5 µl of distilled water were deposited on the wood surface (tangential
face) with a microsyringue and contact angle measured after the total relaxation of the drop
(equilibrium state). Six measurements were performed on each block and the values were
averaged.
2.8. Microscopic analysis
Microscopic observation were performed with an environmental scanning electron
microscope (ESEM Quanta 200) on a small block of 20 x 15 x 15 mm (l, t, r). The transversal
section of the block was prepared with a microtome, repaired with marks and observed under the
scanning microscope. After observation, the sample was heat treated for 4 hours at 250°C under
nitrogen and then, observed again using scanning microscope at the marked points to compare wood
anatomy before and after treatment. This methodology allows to avoid problems of surface
preparation on heat treated samples which were generally very brittle.
75
2.9. Determination of Grevillea robusta chemical composition
2.9.1. Determination of lignin content
Half gram of dry extractive free sawdust (m0) was introduced in 10 ml of 72% concentrated
sulphuric acid at room temperature and stirred for 4 hours. The mixture was then diluted with
240 ml of water and heated for four hours in oil bath at 120°C. The solution was left for 25 minutes
to settle then was filtered through a Buchner, rinsed with hot water at 70°C and dried in autoclave at
70°C to constant weight (m1).
Lignin content (%) = (m1)/(m0) x 100
2.9.2. Determination of holocellulose content
Half gram of dry extract free sawdust (m0) was introduced 2 ml of 15% sodium chlorite and
0.1 ml glacial acetic acid. 2 ml of sodium chlorite and 0.1 ml glacial acetic acid were added after
every one hour for 7 consecutive hours. The solution was filtered through a Buchner filter, rinsed
with hot water and then dried in the oven at 70°C to constant weight (m1).
Holocellulose content (%) = (m1)/(m0) x 100
2.9.3. HPLC analysis
The wood was ground into fine powder and passed through 0.315 mm sieve. Three
hundred and fifty milligrams of the powder was placed in 100 ml flask. Three millilitres of 72%
sulphuric acid was added and the solution placed at ambient temperature for one hour for pre-
hydrolysis. After this, 84 ml of distilled water were added in the mixture and heated in oil-bath at
120°C for four hours to facilitate hydrolysis. After cooling lignin was separated by filtration of
this mixture through a Buchner funnel.
The filtrate containing sugars was transferred into 250 ml flask then diluted with distilled
water to the mark. Twenty millilitres of this solution was neutralized using barium hydroxide
76
(Ba(OH)2. The pH was adjusted to 5. Centrifugation of this solution was performed for
20 minutes at 4000 rev/min. Separation was performed by decanting. Sugar remained in the
solution, then the water was evaporated. These sugars were analyzed by HPLC Photo Array
detector model Waters 2420 ELS. Distilled water was used to dissolve sugars to a concentration of
2 mg/ml. This was injected to the detector using a syringe. Proportions of sugars in the sample was
analyzed by comparing with respective standard solutions.
2.9.4. Acidity titration
Wood samples were treated for different duration of time (30 minutes, 1 hour, 5 hours,
7 hours and 15 hours at a temperature of 250°C) and crushed into powder. Sawdust used for
titration was previously passed through a 115 mesh sieve and dried at 103°C for one night. Acid
value (A) was obtained according to a procedure described in the literature (Matsuda 1987) and
determined as follows: sawdust (250 mg) was mixed with 1ml of 0.1 M HCl in 25 ml of distilled
water and titrated using 0.01 M NaOH with phenolphthalein as indicator. A was obtained by the
following equation :
A (meq/g) = (v – vo) x 10-2 / m
where v is the volume (ml) of 0.01 M NaOH titration solution used for a given wood sample, vo
the volume (ml) of 0.01 M NaOH solution used for neutralizing 1 ml of 0.1 M HCl diluted in 25
ml of distilled water and m the mass of sawdust (g) used for titration.
2.10. Mechanical strength test (MOR and MOE)
Grevillea robusta samples were cut parallel to grain direction and sawn into samples
measuring 120 x 25 x 8 mm (longitudinal, tangential and radial) respectively. These were heat
treated at 240°C varying the treatment time to achieve different weight losses ranging between
2% to 25%. Control samples for each treatment were cut from the same board. The samples were
tested for modulus of Elasticity (MOE) and modulus of rupture (MOR). Comparisons were done
between each control and its thermal treated test samples.
77
3. Results and discussion
3.1 . Heat treatment
Figure 32 represents the evolution of mass loss according to time at different
temperatures.
0
5
10
15
20
25
30
35
40
45
50
0 5 10 15 20 25 30 35
Time (Hours)
Wei
gh
t lo
ss (
%)
(%)
260°C
250°C
240°C
220°C
Figure 32. Variation of the percentage of mass loss with treatment time
Degradation of Grevillea robusta wood was important during the first stage of the thermal
treatment process and becomes less important in a second stage. There was gradual decrease in
mass with time during thermal treatment. The rate of mass loss increased with treatment
temperature for temperatures of 260, 250, 240 and 220°C. The colour of wood darkened with
time of treatment. Treated wood was noted to be brittle as was also observed by Hakkou et al.,
(2005). As reported in the literature on European wood species (Nuopponen et al., 2004 ;
Wikberg and Maunu, 2004 ; Hakkou et al., 2005), these mass losses are probably due to
important degradations of hemicelluloses during the treatment. Mass loss was due to progressive
degradation of hemicelluloses during the treatment due to their amorphous nature (Weiland and
Guyonnet, 2003). A mild pyrolysis of wood in temperature less than 280°C mainly cracks
78
hemicelluloses, but is also reported to begin to modify lignin. By-products of hemicelluloses
condense and polymerize on lignin.
3.2. Improvement of durability
3.2.1. Improvement of durability to fungi
3.2.1.1. Grevillea robusta natural durability
Before evaluation of heat treatment conferred durability, the natural durability of
Grevillea robusta was determined by testing against Coriolus versicolor fungi. Results showed
that Grevillea robusta was very susceptible to attack after three months of exposure in Petri
dishes having fully grown fungi. Samples blocks were fully colonized on the surface as shown in
figure 33.
Figure 33. Grevillea robusta natural durability tested
against Coriolus versicolor
Average weight losses and moisture contents measured at the end of the experiment are
summarized in table 10.
79
Table 10. Natural durability of Grevillea robusta against Coriolus versicolor
Malt agar laboratory tests confirm the low natural durability of Grevillea robusta wood
observed previously with soil test. Even if weight losses observed are lower than those observed
for beech, these latter one are sufficient to consider G. robusta as a non durable species. High
moisture recorded at the end of the experiment could be one of the reason to explain the lower level
of degradation observed. Low durability of Grevillea robusta could be associated to presence of
starch stored on the ray parenchyma cells.
3.2.1.2. Grevillea robusta conferred durability
In a first time, effect of heat treatment time on wood durability was investigated against the
white rot fungus Coriolus versicolor. Results are presented in figure 34 and 35.
Figure 34. Effect of heat treatment time on colonization
of wood by Coriolus versicolor
Sample Weight loss (%) Moisture Content (%)
Control beech 27.0 70.4
G. robusta undried 11.6 98.0
G. robusta dried at 103°C 25 80.2
80
-5
0
5
10
15
20
25
30
Control 30 Min 1 Hour 5 Hours 7 Hours 15Hours
Treatment Time
Wei
gh
t L
oss
%
Figure 35. Effect of heat treatment time on durability to Coriolus versicolor
It was observed that durability increased with treatment time, when Grevillea robusta was
tested against Coriolus versicolor fungi. There is a good correlation between weight loss due to
fungal attack and heat treatment time. These findings are in agreement with the observations of
Hakkou et al., (2006). According to these results, it seems that the duration of pyrolysis and
consequently the mass loss level due to thermal degradation, could be used as a factor to determine
the quality of rectified wood.
According to literature, Hakkou explains the improved durability of heat-treated wood by
four different hypothesis (Hakkou et al., 2006):
- Increase of the hydrophobic character of wood, which limit the sorption of water into the material
hence not favourable for fungal growth.
- Generation of new extractives during heat treatment that act as fungicides.
- Modification of wood polymers leading to non-recognition of the latter by enzymes involved in
fungal degradation.
- Degradation of hemicelluloses, which are one of the main nutritive sources for fungi.
To evaluate the effect of the lower affinity to water of heat treated wood, moisture content
at the end of the biological test (MC) and wettability (θ) have been determined and reported in
81
table 11 with values of mass loss due to heat treatment (ML) and weight loss due to fungal attack
(WL).
Table 11. Effect of time of treatment at 250°C on durability of Grevillea robusta wood
after 3 months exposure to Coriolus versicolor
Sample θ (°) ML (%) MC (%) WL (%) a
Grevillea robusta heat treated 0.5h 113 12 111 3.6 ± 0.5 Grevillea robusta heat treated 1h 116 16 104 2.2 ± 0.6 Grevillea robusta heat treated 5h 113 20.5 88 0
Grevillea robusta heat treated 7h 118 22 78 0
Grevillea robusta heat treated 15h 111 26 49 0
Grevillea robusta control 31 - 97 25.8 ± 11.2 Fagus sylvatica control - - 53 33.5 ± 3.4
a Average value on 4 replicates
The results shows that heat treatment considerably improves wood durability. Different
levels of wood degradation corresponds to different weight losses after varying the treatment time.
Durability against fungus increased with treatment time with optimum resistance reached after 20%
mass losses. At the same time, untreated Grevillea robusta control and beech wood control
virulence were severely degraded by the fungus showing validity of the experiment. Decrease of
wood wettability demonstrated by the higher contact angles measured after treatment has in a first
time no or limited effects on the high moisture content recorded after biological test confirming
results previously reported indicating that hydrophobic character conferred to wood after thermal
treatment wasn’t at the origin of the improvement of durability (Kamdem et al., 2002 ; Gosselink et
al., 2004 ; Hakkou et al., 2005 ; Hakkou et al., 2006). All treated samples have a contact angle θ
ranging between 100 - 120° as shown in figure 36. Treatment time had no effect on wettability
change.
82
θ °
Figure 36. Water drop on heat-treated Grevillea robusta
In the same time, effect of heat treatment was investigated against several wood rotting
fungi, two white rot fungi (Coriolus versicolor and Pycnoporus sanguineus) and four brown rots
(Poria placenta, Coniophora puteana, Gloephyleum trabeum, and Antrodia sp.). Among these
species, two were from tropical origin (Pycnoporus sanguineus and Antrodia sp.). Results are
presented in figures 37.
Coriolus versicolor Pycnoporus sanguineus
Antrodia sp. Coniophora puteana
83
Poria placenta Gloephyleum trabeum
Figure 37. Effect of heat treatment on wood colonization
by different rotting species
Development of the mycelium was important on untreated Grevillea robusta blocks (left
column on each photograph), while heat treated blocks are uncovered or only slightly recovered
by mycelium (middle column on each photograph). Virulence control realized with beech or pine
wood are in all cases strongly colonized by mycelium.
Weight losses due to fungal attack are presented in figure 38.
Coriolus versicolor (Cv), Poria placenta (Pp), Coniophora puteana (Cp), Gloephyleum trabeum Gt), Pycnoporus sanguineus (Ps), Antrodia sp.(Asp)
Figure 38. Weight losses of Grevillea robusta heat treated or not for 7 hours
at 250°C after 3 months exposure to different fungi
-5
0
5
10
15
20
25
30
35
40
Cv Pp Cp Gt Ps Asp
Fungal Species
Wt
Lo
ss (
%)
Heat Treated Sample Control
84
In all cases heat treated Grevillea robusta was resistant to fungal attack. However, these
results must be moderated according to the tested fungus. Indeed, while untreated Grevillea robusta
samples are strongly degraded by Coriolus versicolor and Antrodia sp., heat treated samples show
no degradation showing the efficiency of the treatment. Results are not significant in the case of
other tested fungi, for which untreated samples like heat treated ones were not or only slightly
attacked. Such a behaviour could probably be explained by water logging of the wood blocks hence
not allowing the development of the fungi.
The fact that white rot fungi Coriolus versicolor or Pycnoporus sanguineus, which
normally degrade lignin and polyssacharidic components of wood, were unable to attack
Grevillea robusta wood treated at 250°C for 7 hours, indicates that in addition to polysaccharides
degradation some other factors, like lignin modification or optimal growth conditions, are not
present to allow wood fungal colonization and degradation.
3.2.2. Improvement of durability to termites
3.2.2.1 Heat-treated Grevillea robusta against termites in the laboratory
Results concerning resistance to termites after 28 days in laboratory conditions are
presented in figure 39.
Figure 39. Weight losses of Grevillea robusta heat treated for 7 hours at 250°C
and controls tested against termites in laboratory conditions for 28 days
0
2
4
6
8
10
12
14
1
Wei
ght l
oss
(%)
Untreated Grevillea robusta
Heat treated Grevillea robusta
Untreated Pinus sylvestris
85
Heat treated Grevillea robusta showed insignificant termite attack, while untreated
samples were severely attacked (compare weight losses of less than 1 for control to weight losses
comprised between 11 and 14% for heat treated samples). Untreated Grevillea robusta is more
susceptible to termites than untreated Pinus sylvestris. Survival rate of termites was high in the
controls. Mortality rate was high in the treated samples as the result of feeding on treated wood or
inability to feed on it hence starve to death. After 3 weeks, all the termites in the treated sample
containers were noted to have died.
3.2.2.2. Heat-treated Grevillea robusta against termites in the field
Results obtained under field test conditions are reported in table 12.
Table 12. Weight loss of treated Grevillea robusta and controls
for 4 consecutive months under field conditions
Months 1 2 3 4
Sample Weight
loss (%)
Rating Weight
loss (%)
Rating Weight
loss (%)
Rating Weight
loss (%)
Rating
Heat treated
G. robusta.
0.5 0 0.66 0 0.76 0 0.82 0
Gr. Robusta
control
12.5 1 18.7 2 26.6 3 34 4
P. sylvestris
control
4.4 1 10.7 1 14.2 1 17.3 2
Mean value of 6 replicates
86
There was significant difference between attacks on the heat-treated Grevillea robusta
samples and the controls. Attack on all samples increased with exposure time as presented in
table 12. Pinus sylvestris, the control virulence was attacked but slightly less than G. robusta
control. Untreated Grevillea robusta blocks were strongly degraded, rating quotation ranging 4
for weight loss of 34 %. After four months, heat treated samples remains practically unaffected
by termites.
3.3. Spectoscopic analysis
FTIR and CP/MAS 13C NMR analysis were performed to characterize wood chemical
components.
3.3.1. CP/MAS 13C NMR Analysis
CP/MAS 13C NMR measurements have been used to determine chemical modification of
polymeric wood components after heat treatments (figure 40).
87
Francis_008000fid
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Abs
olut
e In
tens
ity
new control
Francis_004000fid
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Nor
mal
ized
Inte
nsity
1 hour
Francis_005000fid
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Nor
mal
ized
Inte
nsity
7 hours
Heat treated at 250°C
during 1 hour
Heat treated at 250°C
during 7 hour
Figure 39. CP/MAS
13C NMR spectra of Grevillea robusta wood heat treated at 250°C
for different times
Untreated sample indicates classical signals ascribable to main wood polymer
components. Cellulose appears in the region between 60 and 105 ppm. The signals at 72-75 ppm
are assigned to C-2, 3, 5 carbon, the signal at 65 ppm assigned to C-4 carbon and and this at 105
ppm to C-1 hemiacetallic carbon. C-6 signal of cellulose is duplicated due to the presence of
amorphous and crystalline cellulose appearing at 84 and 89 ppm respectively. Signals of
Control
88
hemicelluloses are less obvious due to their overlapping with those of cellulose. The shoulder at
102 ppm on the C-1 signal of cellulose is assigned to hemiacetallic carbon of hemicelluloses.
Acetyl groups of hemicelluloses are detected at 20 and 173 ppm. Methoxyl groups of syringyl
and guaiacyl units of lignin appears at 56 ppm, while aromatic carbons appear between 120 and
160 ppm. Aliphatic carbons of phenylpropane units of lignin are partially masked by
polysaccharidic signals.
Evolution of NMR spectra recorded after heat treatment indicates that the main
modifications appear rapidly after the first hour of treatment. Degradation of hemicelluloses is
particularly obvious. The signal of the methyl of the acetyl groups at 20 ppm, mainly present on
hemicelluloses, disappears totally after heat treatment. The shoulder at 102 ppm on the C-1 signal
of cellulose at 105 ppm decreases as well after heat treatment confirming hemicelluloses
degradation. Similar observations have been reported in the literature on different non durable
European species (Tjeerdsma et al., 1998; Sivonen et al., 2002; Weiland et al., 2003). As reported
in the literature, cellulose cristallinity was more important after heat treatment. The signal at 89 ppm
due to C-4 of crystalline cellulose increases slightly compared to the signal at 82 ppm due to
amorphous cellulose C-4 (Wikberg and Maunu 2004). General behaviour of NMR spectra is
similar to spectra of other wood species reported in the literature excepted for the signal at 30
ppm which is very unusual. Explanation of this signal is due to the presence of important
quantity of lipophilic extractives, approximately 5%, identified after soxhlet extraction with
dichloromethane and 1H NMR analysis (figure 41) to be mainly constituted of n-alkyl and n-
alkenylresorcinols (Ritchie et al., 1965 ; Ridley et al., 1968 ; Cannon et al., 1973). Chemical
shifts of aromatic protons appears as two singlets at 6.19 and 6.26 ppm characteristic respectively
of the proton between the two hydroxyl groups and of the two other protons. Signal at 5.37 ppm
is attributed to vinylic protons of ethylenic double bond. Signals at 2.50 and 2.04 ppm are
characteristic of benzylic and allylic protons, while signals at 1.59, 1.28 and 0.91 correspond to
methylene and methyl groups of fatty alkyl chain. Proton of phenolic hydroxyl group appears as a
singlet at 7.28 ppm.
89
ppm (f1) 1.02.03.04.05.06.07.0
7,28
6,26
6,19
5,37
2,50
2,04
1,59
1,28
0,91
Figure 41. 1H NMR spectra of dichloromethane extractives of Grevillea robusta wood
Changes in the lignin signals are less significant indicating its higher stability. The
appearance of new broad aromatic or alkenic resonances (110 to 135 ppm) and aliphatic
resonances (10 to 50 ppm) after heat treatment indicate structural modifications of wood
polymers due probably to a beginning of destruction of the amorphous cellulose structure. Such
modifications, previously reported during cellulose thermal treatment (Zawadzki and
Wisniewski, 2002), could be at the origin of the increase of cristallinity. Analysis of the sample
heat treated for 15 hours indicated important degradation of the wood polymers even if the
measured mass loss due to thermal degradation was quite similar to that observed after 7 hours
(compare 22% to 26%) confirming cellulose degradation previously evocated (figure 42).
R
HO
HO
R = -(CH2)nCH3 or -(CH2)nCH=CH(CH2)n'CH3
90
Francis_007000fid
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Nor
mal
ized
Inte
nsity
15 hours
Figure 42. CP/MAS 13C NMR spectra of Grevillea robusta heat treated
at 250°C for 15 hours
Assignment of these new signals (between 110 to 135 ppm and 10 to 50 ppm) could be probably
attributed to a beginning of formation of carbonaceous material within the wood polymers.
Indeed, 13C NMR studies reported during cellulose thermal treatment performed in the range of
temperature from 300 to 600°C indicates quite similar observations for samples treated at 300°C
(Zawadski et Wisniewski 2002). Thermal degradation of cellulose leads in a first time to the
apparition of new signals ascribable to aliphatic carbons (10-50 ppm), aromatic or alkenic
carbons (110-160 ppm) and carbonyl groups (205 ppm). At this temperature, cellulose is reported
to undergo depolymerization by transglycosylation to form after dehydration reactions
anhydromonosaccharides like 1,6-anhydro-β-D-glucopyranose (levoglucosan) and 1,6-anhydro-
β-D-glucofuranose. These products are involved in subsequent reactions leading to char
formation (Rowell et Le Van-Green. 2005). These types of structural changes have also been
reported by others (Plaisantin et al. 2006).
91
3.3.2. FTIR Analysis
FTIR analysis is presented in figure 43.
4000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.0
10.00
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0
65.00
cm-1
%T
4000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.0
21.53
25.0
30.0
35.0
40.0
45.0
50.0
54.60
cm-1
%T
4000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.0
2.29
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
47.40
cm-1
%T
Heat treated at 250°C
during 1 hour
Control
Heat treated at 250°C
during 7 hours
3350
1600 1510
1450
1058
3350
1600
1510
1450
1058
3350
1600
1510
1450
1058
Figure 43. FTIR spectra of Grevillea robusta wood after thermal treatment
92
Behaviour of wood components is more difficult to interpret due to the similarities of the
spectra before and after treatments. However some small modifications can be observed
indicating the same tendencies as previously observed by NMR. The O-H absorption band at
3350 cm-1 decreased compared to all other bands. Characteristic C-O absorption band of
polyssacharidic components at 1058 cm-1 corresponding to C-O stretching vibrations decreases
slightly comparatively to lignin characteristic band at 1600, 1510 and 1450 cm-1 confirming the
degradation of hemicelluloses.
3.4. Chemical composition of Grevillea robusta wood
3.4.1. Extractives content
Table 13 shows the percentage of extractives from Grevillea robusta by soxhlet method
using different solvents. The measurement was done by direct method (DM) and indirect method
(IM) for all the samples.
Table 13. Percentage extract from Grevillea robusta by soxhlet method
using dichloromethane, toluene/ethanol and water
Sample Heartwood
DM (%) IM (%)
Dichloromethane 5.2 5.9
Toluene/Ethanol 7.4 8.5
Water 3.4 3.5
Grevillea robusta extractives content is relatively low compared to content generally
observed in durable tropical hardwood species. Extracts contains higher quantity of non-polar
components like alkyl and alkenyl resorcinol compared to polar components obtained after water
extraction.
3.4.2. Lignin and holocellulose content
Lignin and holocellulose contents obtained after acidic hydrolysis of polysaccharides or
delignification with sodium chlorite are presented in table 14.
93
Table 14. Chemical composition of Grevillea robusta
Holocellulose (%) Lignin (%)
Measured 2 Estimated
3 Measured
4 Estimated
5
70.3 76.5 23.5 29.7
2 holocellulose content measured after delignification with sodium chorite 3 holocellulose content estimated by difference with lignin content 4 lignin content measured after acidic hydrolysis of polysaccharides 5 lignin content estimated by difference with holocellulose content
Results show that Grevillea robusta possess classical contents of holocelluloses and
lignin.
3.4.3. Acidity titration
Wood decaying basidiomycetes have a pH optima between 3 to 6 (Keilich et al. 1970;
Highley 1975; Zabel and Morrel 1992) with brown rot fungi having the lowest optima around 3.
Considering the important degradation of hemicelluloses it seems interesting to
investigate the acidity of wood in relation with degradation of wood polymers and durability.
Indeed, degradation of hemicelluloses could reduce the acidity of wood due to the decomposition
of uronic acids present in their structure. On the other hand, formation of acetic acid could
increase the pH if this latter one remains in the wood.
Titration was performed using an automated burette of high precision. For each titration, the
volume of NaOH used to neutralize 250mg of sawdust in suspension in 25 ml of distilled water and
1 ml of 0.1N HCl was displayed on a computer connected to the burette as recorded in table 15.
94
Table 15. Titration volume of 0.01N NaOH used to neutralize acidity present in wood
Titration Av. Vol. of NaOH (ml) ML after heat treatment (%)
“1 ml 1N HCl + water” 9.7
Control “Not Treated” 13.7
Control “Water extracted” 11.6
Heat TT (30min) 10.7 12
Heat TT (1hr) 10.2 16
Heat TT (5hr) 10.0 20.5
Heat TT (7hr) 9.7 22
Heat TT (15hr) 9.5 26.5
Titration volume of NaOH reduced gradually with the increase in treatment time as
illustrated in figure 43. This may be explained by disappearance of carboxylic acid functions present
on hemicelluloses as they decomposed during the treatment. Measurements realized on control and
water extracted control indicates that a part of the acidity measured is not due to water soluble
compounds, but to polymeric components of wood.
According to these results, it seems that acidity initially present in wood decreases with heat
treatment time to zero for treatment time of more than 7 hours (figure 44).
0
0,02
0,04
0,06
0,08
0,1
0,12
0,14
0,16
0,18
0 2 4 6 8 10 12 14 16
time (hour)
A (
meq
./g)
Figure 44. Evolution of wood acidity with treatment time
95
3.4.4. Sugar composition
Acidic hydrolysis of wood polysaccharides allows the determination of Klason lignin
and HPLC determination of simple sugar initially present in wood (table 16).
Table 16. HPLC determination of sugars in untreated and heat treated Grevillea robusta
starting from 350 mg of sawdust
Grevillea robusta
treatment at 250°C Control
1 hour
ML = 16%
5 hours
ML = 20.6%
7 hours
ML = 22%
15 hours
ML = 26.5%
Weight loss after
Treatment (%) 0 16 20.5 22 26
Lignin (mg) 84.6 126.6 137.0 158.1 167.5
Glucose (mg) 186.2 206.9 180.5 168.7 176.0
Xylose (mg) 45.1 9.6 7.0 9.5 0.0
Galactose(mg) 8.4 3.3 0.0 0.0 0.0
Rhamnose(mg) 2.66 0.6 0.0 0.0 0.0
Arabinose(mg) 0.0 0.0 0.0 0.0 0.0
Results shows that lignin content increased with treatment time. The amounts of simple
sugars which make-up hemicelluloses, reduce with treatment time and were exhausted from wood
after 7 hours of thermal treatment. Decomposition of simple sugars and increase of lignin content
may explain previous observation of Grevillea robusta durability against fungi after 7 hours of
thermal treatment. Glucose which forms celluloses was resistant to thermal degradation as shown by
low weight loss even after 15 hours of thermal treatment.
3.5. Microscopic analysis
Microscopic analysis of Grevillea robusta heartwood has been performed before and after heat
treatment to evaluate effect of polymers degradation on wood anatomical structure. To avoid
artifacts due to sample preparation, especially in the case of heat treated wood which is very
brittle and difficult to cut, observations have been performed on the same piece of wood.
96
Figure 44 represents transverse section of Grevillea robusta wood at different magnifications.
Figure 44. Cross section of Grevillea robusta wood
The wood present successive rings appearing like concentric bands which can be
associated with the dry and wet periods. This is a ring-porous wood, with tangential bands of
large, porous vessels surrounded by high amounts of parenchyma. Fibres occupy the wider gaps
between the rings. Large and numerous rays are present. Vessels contain various amounts of
deposits.
Figure 45 present photographs taken at the same place of the sample of Grevillea robusta
wood before and after heat treatment under nitrogen for 4 hours at 250°C (ML = 20.1%).
97
Before heat treatment After heat treatment
Figure 45. Cross sections of Grevillea robusta wood before and after heat treatment
Microscopic analysis indicates that anatomical structure of wood was only slightly
affected during treatment. Vessels, fibers, parenchyma and rays are still obvious after heat
treatment. The main differences was the presence of important quantities of extractives deposited
in the vessels, which disappear after thermal treatment (white arrows indicating extractives
deposited in the vessels). The diameter of vessels remained unchanged after treatment.
Heat treatment result also to some shrinkage. Due to the anisotropy of wood, the
shrinkage depends on the orientation. Tangential shrinkage was approximately 6 percent, while
radial shrinkage was approximately 2 percent.
3.6. Mechanical properties
Elasticity implies that deformations produced by low stress are recoverable after the loads
are removed. When loaded to higher stress levels, plastic deformation or failure occur. The three
moduli of elasticity EL, ER and ET are along longitudinal, radial and tangential axis of wood
98
respectively. The modulus of elasticity determined from bending, EL, is commonly used and was
considered modulus of elasticity (MOE) during this test.
Modulus of rupture (MOR) reflects the maximum capacity of a member in bending and is
proportional to maximum moment borne by the specimen as presented in figure 46.
0,0
0,2
0,4
0,6
0,8
1,0
1,2
1,4
0 1 2 3 4
displacement
forc
e
Figure 46. Force against displacement in MOE and MOR test
Heating wood to high temperatures changes its physical and chemical properties.
As a result, dimensional stability is increased due to reduced moisture intake and decay
resistance is enhanced. Strength and brittleness of wood decrease depending on the
species, anatomical features and treatment methods (Viitaniemi, 1993). Indeed, it is
known that an important drawback of heat treated wood is its mechanical brittleness
(Santos 2000, Mouras et al. 2002, Unsal and Ayrilmis 2005, Yildiz et al. 2006).
For these reasons, heat treated wood is not recommended for use in load-bearing
constructions, but can find valuable applications for furnitures, wall, ceiling, roofing,
flooring and fencing.
Results obtained in the case of Grevillea robusta are presented in figures 47 and 48.
MOE
MOR
99
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
Control 240°C, WL = 12.4%
240°C, WL = 15.7%
240°C, WL = 19.2%
240°C, WL = 25%
220°C, WL = 16.5%
MOR (kPa)
MOE (Mpa)
Figure 47. MOE and MOR against weight loss after thermal treatment at 240 and 220°C
0
10
20
30
40
50
60
70
80
240°C, WL = 12.4%
240°C, WL = 15.7%
240°C, WL = 19.2%
240°C, WL = 25%
220°C, WL = 16.5%
Red
uctio
n of
mec
hani
cal p
rope
rtie
s (%
) MOR (kPa)
MOE (Mpa)
Figure 48. Reduction (%) in MOE and MOR with weight loss
after thermal treatment at 220 and 240°C
100
Results showed that modulus of elasticity (MOE) decreased insignificantly for Grevillea
robusta samples having weight loss of less than 12% after thermal treatment. There was
significant reduction in MOE for heat-treated Grevillea robusta with weight loss of 16% and
above. The reduction progressed with increase in weight loss due to thermal treatment. Results
also showed that treatment temperature was not the main factor contributing to reduction in
MOR. Weight loss after thermal treatment, regardless of treatment temperature seems to be a
plausible explanation. Samples treated at 220°C with a weight loss of 16% showed similar MOR
values as those treated at 240°C with same weight loss.
Reduction of modulus of rupture (MOR) is more significant. Independent of treatment
conditions MOR is reduced by at least half of its initial value indicating an important decrease of
the mechanical properties. The modulus of rupture decreased with increase in weight loss of
Grevillea robusta after thermal treatment as presented in figure 36. The decrease in strength is
mainly due to the depolymerization reactions of wood polymers. Changes in the amount of
hemicelluloses play an important role in strength properties of wood at high temperatures (Hillis,
1984 ; Yildiz et al., 2006).
101
4. Conclusion
This study shows that heat treatment of Grevillea robusta increases its durability against
basidiomycetes and termites. The decay resistance depends of the treatment condition and is total
for treatments performed at 250°C for 5 hours. Improvement of heat treatment conditions could
probably lead to smaller duration and temperature.
Wettability changes observed after the heat-treatment cannot be used to explain wood
enhanced durability. Similarly to studies reported on non durable European wood species,
chemical modifications seems to be the most plausible hypothesis to explain wood durability
improvement. Mass losses, 13C MAS NMR and FTIR analysis indicate important degradation of
hemicelluloses. Moreover, wood chemical analysis and NMR analysis indicate the presence of
degradation products within the wood polymeric components due to a beginning of degradation
of cellulose. Another possibility could be the decrease of wood acidity creating unfavourable
conditions for the development of wood rotting basidiomycetes.
Microscopic analysis indicate that wood anatomy was slightly affected and that some
shrinkage occurred after treatment. The main drawback of the treatment concerns the reduction of
mechanical properties. For these reasons, heat treated wood is not recommended for use in load-
bearing constructions, but can find valuable applications for furnitures, wall, ceiling, roofing,
flooring and fencing.
All these results show that heat treatment could be a valuable alternative to toxic
chemicals used actually to treat Grevillea robusta wood in Kenya. This is confirmed by improved
durability against fungi and termites in the laboratory and in the field. However, our work is only
a preliminary feasibility study. It should be performed on a larger scale, if possible with industrial
partnership to evaluate real possibility of the method.
102
General conclusion and recommendations
The objectives of the first part of this work were achieved. Durability tests of Prunus
africana wood species against fungi and termites was correlated to extractives, more so
dichloromethane and water extracts. Analysis of extracts believed to have fungicidal properties is
encouraging for understanding the reasons of exceptional durability of Prunus africana in Kenya
and correlated to the presence, like in bark, of phytosterols. These results justify further
researches on the separation, identification and characterization of fungicidal and termicidal
properties of Prunus africana dichloromethane extracts allowing the development of new family
of biocides based analogies with natural compounds.
The second part of this work showed that Grevillea robusta is treatable to retention levels
that meet minimum requirements for various use conditions including ground contact or
permanent submersion in fresh and sea water using sap displacement method. The positive linear
relationship between retention, penetration and solution strength means that the desired retention
and penetration of preservative can be achieved by varying the solution strength. Preservative
penetration is dependent of the height of the pole and inversely proportional to the distance from
the end dipped in preservative. Undebarked specimens tend to retain less preservative, while the
debarked ones yield greater penetration. Debarked G. robusta poles can achieve retention
required for wood permanently submerged in fresh water at 2 % solution strength and for marine
water exposure at solution strength of 4% and 6%. Undebarked wood can be treated to retention
levels required for ground contact using 2% solution strength and for fresh water and marine
exposure using 4% and 6% solution concentrations.
The study showed also that thermal treatment of Grevillea robusta allowed improvement
of some of wood properties. Heat treatment of Grevillea robusta increases its durability against
basidiomycetes and termites. Decay resistance depends on the treatment condition and total
resistance is attained for treatments performed at 250°C for 5 hours. Wettability changes
observed after the heat-treatment cannot offer explanation to enhanced wood durability. Similarly
to studies reported on non durable European wood species, chemical modifications seem to be the
most plausible hypothesis to explain wood durability improvement. Mass losses, 13C MAS NMR
103
and FTIR analysis indicate important degradation of hemicelluloses. Moreover, NMR analysis
indicates the presence of degraded products within the wood polymeric components due to a
beginning of degradation of cellulose. Microscopic analysis indicates that wood anatomy was
slightly affected and that some shrinkage occurred after treatment. These results show that heat
treatment could be a valuable alternative for Grevillea robusta for furniture, wall, ceiling,
roofing, flooring and fencing purposes, hence conservation of forest resources in Kenya.
Three articles have published from the preceding results :
- Impregnability of Grevillea robusta using sap displacement method. F. Mburu, F. Muisu,
P. Sirma, P. Gérardin. Bois et Forêts des Tropiques, N° 286 (4), 65-72, 2005 (appendix 1)
- Evaluation of thermally modified Grevillea robusta heartwood as an alternative to
shortage of wood resource in Kenya : characterisation of physicochemical properties and
improvement of bio-resistance. F. Mburu, S. Dumarçay, F. Huber, M. Pétrissans,
P.Gérardin, to be published in Bioresource Technology (appendix 2)
- Evidence of fungicidal and termicidal properties of Prunus africana heartwood
extractives. F. Mburu, S. Dumarçay, P. Gérardin, to be published in Holzforschung
(appenedix 3)
Recommendations
1) Other wood treatment techniques should be explored and tested using Grevillea robusta
wood species due to its extensive usage by local farmers in Kenya.
2) Thermal treatment technique should be tested on other wood species like Acacia meansii
which is also not durable and widely used in Kenya.
3) Development of heat treatment of Grevillea robusta with an industrial plant could be of
valuable interest to evaluate potentiality of this process in real conditions in Kenya.
104
Conclusion générale et perspectives
Les objectifs de la première partie de ce travail ont été atteints. La durabilité naturelle du
bois de cœur de Prunus africana vis à vis des champignons et des termites est fortement liée à la
présence d’extractibles et tout particulièrement aux extraits solubilisés dans l’eau ou le
dichlorométhane. L’analyse des extraits responsables des propriétés fongicides permet de mieux
comprendre les raisons de la durabilité exceptionnelle de cette essence qui semble corrélée,
comme dans l’écorce, à la présence de phytostérols. Ces résultats justifient d’autres études sur la
séparation, l’identification et la caractérisation des propriétés fongicides et termiticides des
extraits dichlorométhane du Prunus africana permettant le développement de nouvelles familles
de biocides mimétiques de produits naturels.
La seconde partie de ce travail a permis de montrer que le Grevillea robusta était
imprégnable en utilisant une méthode de déplacement de sève avec des quantités de biocides
suffisantes pour atteindre les différentes classes de risques incluant les utilisations au contact du
sol ou dans l’eau douce ou l’eau de mer.
La relation linéaire existant entre la rétention, la pénétration et la concentration de la
solution de traitement permet d’envisager l’imprégnation de différentes quantités de biocides en
fonction de la concentration initiale. La pénétration du produit de préservation est fonction de la
hauteur du poteau et inversement proportionnelle à la distance entre le bas et haut du poteau. Les
poteaux non écorcés retiennent plus de produit de préservation, alors que ceux écorcés conduisent
à une meilleure pénétration. Les poteaux écorcés peuvent être traités à des taux de rétention
suffisants pour une utilisation au contact de l’eau douce à partir d’une solution à 2% et de 4 et 6%
pour des utilisation en eau de mer. Les poteaux non écorcés peuvent être utilisés au contact du sol
pour des solutions d’imprégnation de 2% et au contact de l’eau douce ou de l’eau de mer à partir
de 4 et 6%.
L’étude montre également que le traitement thermique permet d’améliorer certaines
propriétés du Grevillea robusta. Le traitement thermique augmente la durabilité du bois face aux
basidiomycetes et aux termites. La résistance aux agents de pourritures dépend des conditions de
105
traitement et est totale pour des traitement réalisés pendant 5heures à 250°C. Les changements de
mouillabilité observés après le traitement thermique ne permettent pas d’expliquer
l’augmentation de durabilité. De la même manière que lors d’études réalisées sur des essences
européennes non durables, les modifications chimiques intervenant au cours du traitement
semblent être l’hypothèse la plus probable pour expliquer l’augmentation de durabilité. Les pertes
de masse, les analyses RMN
13C MAS et FTIR indiquent une forte dégradation des
hémicelluloses. De plus, l’analyse RMN permet de mettre en évidence la présence de produits de
dégradation à l’intérieur de la structure du bois en plus des polymères initialement présents
résultant de la dégradation des polysaccharides avec un début de dégradation de la cellulose. Les
analyses microscopiques montrent que l’anatomie du bois et peu affectée au cours du traitement
et que ce dernier conduit également à un léger retrait. Ces résultats montrent que le traitement
thermique peut être une alternative intéressante pour augmenter la durabilité du Grevillea robusta
pour des applications telles que la fabrication de meubles, de murs, de plafond, de toitures ou de
clôtures, permettant ainsi de préserver les ressources forestières naturelles du Kenya.
Les résultats obtenus lors de ce travail ont permis la publication de trois articles :
- Impregnability of Grevillea robusta using sap displacement method. F. Mburu, F. Muisu,
P. Sirma, P. Gérardin. Bois et Forêts des Tropiques, N° 286 (4), 65-72, 2005 (annexe 1)
- Evaluation of thermally modified Grevillea robusta heartwood as an alternative to
shortage of wood resource in Kenya: characterisation of physicochemical properties and
improvement of bio-resistance. F. Mburu, S. Dumarçay, F. Huber, M. Petrissans,
P.Gérardin, à paraître dans Bioresource Technology (annexe 2)
- Evidence of fungicidal and termicidal properties of Prunus africana heartwood
extractives. F. Mburu, S. Dumarçay, P. Gérardin, à paraître dans Holzforschung (annexe
3)
.
Perspectives
1) d’autres techniques de traitement devraient être étudiées du fait de l’utilisation intensive
du Grevillea robusta au Kenya
106
2) le traitement thermique devrait être appliqué à d’autres essences non durables du Kenya
3) le développement du traitement thermique du Grevillea robusta devrait être réalisé de
manière industrielle de façon à réellement évaluer les potentialités de la méthode dans des
conditions réelles d’utilisation au Kenya.
107
References
Adaskaveg J. E. (1992). Defense mechanism in leaves and fruits of trees to fungal infection. In:
Defense mechanisms of woody plants against fungi. Ed Blanchette and Briggs, Springer
Verlag. Berlin, 207-245.
Alén R., Kotilainen R. and Zaman A. (2002). Thermochemical behaviour of Norway spruce (Picea
abies) at 180-225°C. Wood Science Technology, 36, 163-171.
Anke H., Roland W. and Weber S. (2006). White-rots, chlorine and the environment - a
tale of many twists. Mycologist, 20(3), 83-89.
Arango R.A., Green F., Hintz K., Lebow P. K. and Miller R. B. (2006). Natural durability of
tropical and native woods against termite damage by Reticulitermes flavipes (Kollar).
International Biodeterioration & Biodegradation, 57(3), 146-150.
Assersson R. and Bergman O. (1971). Preservation methods for chips used in pulping. In:
"Biodeterioration of materials". Walter and Van der Plas (eds), Vol. 2, part XI, Applied Science
publishers, New York. pp. 380.
Bagine R.K.N. (1992). Problems in identification and control of termites. Termite research and
control proceedings. National Museum of Kenya (Nairobi) 1994.
Barlet A., Albrecht J. and Aubert A. (1990). Efficacy of Pygeum africanum extract in the medical
therapy of urination disorders due to benign prostatic hyperplasia: evaluation of objective and
subjective parameters. A placebocontrolled double blind multicenter study. Wien Klin
Wochenschr, 102, 667-673.
Bassi P., Artibani W., Deluca V, Zattoni F. and Lembo A. (1987). Standardized extract of Pygeum
africanum in the treatment of benign prostatic hypertophy. Minerva Urology Nefrology, 39, 45-
50.
Baya M., Soulounganga P., Gelhaye E., Gerardin P. (2001). Fungal activity of β-thujaplicin analogues. Pest manag. Sci., 57, 833-838.
Becker H., Scher, J.M., Speakman, J. B. Zapp, J. (2005). Bioactivity guided isolation of
antimocrobial compounds From Lythrum salicaria. Fitoterapia, 76, 580-584.
Blanchette R.A., Farrel R.L., Burnes T.A., Wendler PA Zimmerman W., Brush T.S. and Snyder
R.A. (1992). Biological control of pitch in pulp and paper production by Ophiostoma
piliferum. Tappi J., 75,102-106.
Block S.S. (1953). Humidity requirements for mould growth. Applied Microbiology, 6, 287-293.
Bombardelli E. and Morazzoni P. (1997). Prunus africana (Hook. f.) Kalkm Fitoterapia, 68, 205-
218.
108
Breza J., Drurny O., Borowka A., Hanus T., Petrik R., Blane G. and Chadha-Boreham H. (1998).
Efficacy and adaptability of Tedenan (Pygeum africanum) in the treatment of benign prostatic
hyperplasia (BPH): a multicenter trial in central Europe. Current Medical Research Opinion,
14, 127-139.
Brisson A., Gharibian S., Eagen R., Leclerc D.F. and Breuil C. (1996). Localization and
characterization of the melanin granules produced by the sap-staining fungus Ophiostoma
piceae. Mater Org., 30, 23-32.
Cannon J.R., Chow P. W., Fuller M.W., Amilton B.H., Metcalf B.W., Power A.J. (1973).
Phenolic constituents of Grevillea robusta (Proteaceae). Structure of robustol, a novel macro
cyclic phenol, Australian Journal of Chemistry, 26(10), 2257-2275.
Carlile M. and Watkinson S.C. (1994). The Fungi. Academic Press Inc, San Diego, USA.
Cartwright, K.ST.G. and Findlay, W.P.K. (1958). Decay of timber and its prevention. 2nd. Ed. Her
Marjesty Stationary office, London.
Catalano S., Ferretti M., Marsili A., Morelli I. (1984). New constituents of Prunus africana bark
extract. Journal of Natural Products, 47(5) 910.
Celimene C. C., Micales J. A. Ferge L. and Young R. (1999). Efficacy of pinosylvins against white
rot and Brown rot fungi. Holzforschung, 53(5), 491-497.
Clausen C.A. (1996). Bacterial associations with decaying wood: a review. Int Biodeterior
Biodegrad, 37, 101-107.
Collins N.M. and Webber J.F. (1987). Insects fungus interaction. Harcourt Brace Jovanovich
publishers, London. pp. 334.
Cunningham A.B. Mbenkum F.T. (1993). Sustainability of harvesting Prunus africana bark in
Cameroon: a medicinal plant in international trade. People and Plants Working Paper 2, pg 1-
28, UNESCO, Paris.
Cunningham M., Cunningham, A.B. Schipmann, U. (1997). Trade in Prunus africana and the
implementation of CITES. Results of the R+D Project 808 05 080. German Federal Agency for
Nature Conservation.
Deacon J.W. (1997). Modern Mycology. 3rd Ed, Blackwell Sciences Ltd, Edinburgh, UK. Pp
199-203.
Deepak M. and Handa S.S. (2000). Antiinflamatory Activity and chemical composition of extract
of Verbena officinalis. Phytother 14, 463-465. (Short communication).
Deon G., Chadenson M., Hauteville M. (1980). Influence des extraits de bois sur sa resistance à
la pourriture. Revue Bois et forêts des Tropiques, No 191, 75-90.
109
Dickinson D.J. and Levy J.F. (1979). Mechanisms of decay and its prevention. Record of the 1979
British Wood Preservers Association annual convention 1979.
Dorado J., Beek T.A., Claassen F.W. and Sierra-Alvarez R. (2001). Degradation of lipophilic
wood extractive constituents in Pinus Sylvestris by the white-rot fungi Bjerkandea sp. and
Trametes versicolor. Wood Science and Technology, 35, 117-125.
Dufour B., Choquenet C., Revol M., Faure G., and Jorest R. (1984). Controlled study of the effects
of Pygeum africanum extract on the functional symptoms of prostatic adenoma. Ann Urol
(Paris) 18, 193-195.
Eaton R.A. and Hale M.D.C. (1993). Wood decay, pests and protection. Chapman & Hall, 2-6
Boundary Row, London.
Eriksson K.E.L, Blanchette R.A. and Ander P. (1990). Microbial and enzymatic degradation of
wood and wood components. Springer-Verlag, Berlin, Germany.
Escalente M.A., Santecchia, C.B., Lopez S.N., Gattuso M.A., Ravelo A.G., Monache F.D.,
Sierra M.G., Zacchino S.A. (2002). Isolation of antifungal saponins from Phytolacca
tetramera, an Argentinean species in citric risk. Journal of Ethnnopharmacology, 82, 29-34.
FAO (1986). Food and Agriculture Organization; Wood Preservation Manual.
Fengel D., Wegener G. (1984). Chemical analysis and composition of wood in Wood Chemistry,
Ultra structure, Reaction, Walter de Gruyter & Co, Berlin, Germany.
Flodin K. and Fries N. (1978). Studies on volatile compound from Pinus sylvestris and their
effect on wood decomposing fungi, II: Effects of some volatile compounds on fungal
growth. Eur. J. For. Pathology, 8, 300-310.
Floyd A.G. (1989). Climates of the natural range. In: Grevillea robusta in Forestry and Agro
forestry (Ed). C. E. Harwood, ICRAF, Nairobi, Kenya.
Fourneau, C., Hocquemiller, R., Cavé, A. (1996). Triterpenes from Prunus africana bark.
Phytochemistry, 42(5), 1387-1389.
Fowlie I.M. and Sheard L. (1983). Developments in the Use of Homegrown Spruce Poles for Use
as Overhead Line Supports. In: Wood: Decay, Pests and Protection. (Eds.) R.A. Eaton &
M.D.C. Hale. Chapman & Hall, London.
Freas A. (1982). Evaluation, maintenance and upgrading of wood structures: A guide and
commentary. American Society of Civil Engineers, New York, USA.
Gérardin P., Neya B., Dumarcsay S., Pétrissans M., Serraj M., Huber F. (2004). Contribution of
gums to natural durability of Prosopis africana heartwood. Holzforschung, 58(1), 39-44.
110
Gitonga W. (1995). Comparison of chemical and cultural treatments against termites in western
Kenya. Proceedings of Second Regional Workshop on termite. research and control. Kenya Forestry Research
Gosselink R.J.A., Krosse A.M.A., van der Putten J.C., van der Kolk J.C., de Klerk-Engels B., van
Dam J.E.G. (2004). Wood preservation by low-temperature carbonisation, Industrial Crops
and Products, 19, 3-12.
Grace J. K. (1996). Susceptibility of compressed fiber to termite attack. Forest Products journal,
46(9), 76-78.
Green F., Kuster T. and Highley T. (1997). Targeted Inhibition of Wood Decay (Using Everything
but Kitchen Sink). International Research Group Paper IRG/WP-10203.
Hakkou M., Pétrissans M., El Bakali I., Gerardin P., Zoulalian, A. (2005). Wettability changes
and mass loss during heat treatment of wood. Holzforschung, 59(1), 35-37.
Hakkou M., Pétrissans M., Gérardin P. and Zoulalian A. (2006). Investigation of the reasons for
fungal durability of heat-treated beech wood. Polymer degradation and Stability, 91(2), 393-
397.
Hale M.D.C. and Eaton R. A. (1986). Soft rot cavity widening; a kinetic approach. Proceeding of the
Royal society of London B227, pp.223.
Hart J.H. (1982). Variation inherent decay resistance of black Walnut. US For Service Tech. Rep.
N C., 74.
Harwood C.E. and Booth T. H. (1992). Status of Grevillea robusta in Forestry and Agro forestry.
In: Grevillea robusta in Forestry and Agro forestry. (Ed.) C.E. Harwood. ICRAF, Nairobi.
Haupt M., Leithoff H., Meier D., Puls J., Richter H.G., Faix O. (2003). Heartwood extractives
and natural durability of plantation-grown teakwood (Tectona grandis L.) - A case study.
Holz als Roh und Werstoff, 61(6), 473-474.
Highley T.L. (1975) Properties of cellulases of two brown-rot and two white-rot fungi. Wood
Fiber, 6, 275-281.
Hickin N. E. (1971). Termites a world problem. Hutchinson and Co. Publishers Ltd. London. pp.
375.
Hickin N. E. (1975). The insect factor in wood decay. Anchor press, London. pp. 383.
Hill C. (2005). The Use of timber in the twenty-first century. In Wood Modification, Chemical
Thermal and other Processes, Wiley Series in Renewable Resources, ISBN-10 0-470-02172-
9, 1-18.
Hillis W.E. (1984). High temperature and chemical effects on wood stability. Wood Science and
Technology, 18, 281-293.
111
Hillis W.E. and Sumimoto M. (1989). Effect of extractives on pulping. In J.W. Rowe, Ed,
Natural Products of Wood Plants II, Springer-Verlag, New York pp 880-920.
Hitimana J. (2000). Variations in structure, composition and regeneration within the Mt Elgon
moist lower montane forest, Kenya, MPhil Thesis, Moi University, Eldoret, Kenya.
Holt D.M., Garreth J.E.B., and Furtado E.J (1979). Bacterial breakdown of wood in aquatic habitats.
Record of the 1979 Annual Convention of the British Wood Preservers Association.
ICRAF. (1992). A selection of useful trees and shrubs for Kenya. Their identification, propagation,
and management for use by farming and pastoral communities, ICRAF, Nairobi, Kenya.
ICRAF. (1999). Leading the path to prosperity, ICRAF, Nairobi, Kenya.
ICRAF. (2000). Calender, ICRAF, Nairobi, Kenya.
Inamori Y., Sakagami Y., Morita Y., Shibata M., Sigiura M., Kumeda Y., Okabe T., Tsujibo H.,
Ishida N. (2000). Antifungal activity of hinokitiol-related compounds on wood-rotting fungi
and their insecticidal activities. Biol. Pharm. Bull., 8, 995-997.
Johnson B.R., Gjovik L.R., Roth H.G. (1973). Single and dual-treated panels in a semi-tropical
harbour. Preservative and retention variables and performance. American wood preservers
Assoc. Proc., 69, 207-215.
Käärik A. (1980). Fungi causing sap stain in wood. IRG/WP 199, The International Research
Group on Wood Preservation, IRG Secretariat Stockholm, Sweden.
Kamdem D.N., Pizzi A., Jermannaud A. (2002). Durability of heat-treated wood. Holz als Roh-und
Werkstoff, 60(1), 1-6.
Kamweti D.M. (1992). Growth and utilization of Grevillea robusta around Mt Kenya: In Grevillea
robusta in Forestry and Agro forestry (Ed) C. E. Harwood, ICRAF, Nairobi.
Keilich G., Bailey P.J., Liese M. (1970). Enzymatic degradation of cellulose derivatives and
hemicelluloses in relation to the fungal decay of wood. Wood Science and Technology, 4,
273-283.
KFMP (1994). Kenya Forestry Master plan Development Programmes Ministry of Environment and
Natural Resources, Government of Kenya, 1-208.
Kinyanjui T., Gitu P.M., Kamau G.N. (2000). Potential antitermite compounds from Juniperus
procera extracts. Chemosphere, 41(7), 1071-1074.
Kokwaro J.O. (1976). Medicinal Plants of East Africa: East African literature Bureau, Nairobi,
Kenya. Pp 384.
112
Kollman F.F.P. and Cote W.A. Jnr. (1984). Principles of Wood science and Technology. Vol. 1.
Solid wood. Syracuse University Press, New York.
Lee K.E. and Wood T.G. (1971). Termites and Soil. Academic Press, New York.
Letouzey R. (1978). Rosaceas. Flore du Cameroun, laboratoire du phanerogamie, Museum National
d’Histoire Naturelle, Paris, 20, pp 195-244.
Levya A., Dimmel D.R., and Pullman D.S. (1998) Teak extract as catalyst for the pulping of loblolly
pine. Tappi journal, 81(5), 237-240.
Lopez-Carbonell M., Moret A., Nadal M. (1998) Changes in cell ultra structure and zeatin riboside
concentrations in Hedera helix, Pelargonium zonale, Prunus avium and Rubus ulmifollius
leaves infected by fungi. Plant Disease journal, 82(8), 914-918.
Mc Coy Hill M. (1965). Travaux du Centre de Recherches et d’études. Océanographiques, 6(1-2-3-
4), 389-396.
Matsuda, H. (1987) Preparation and utilization of esterified wood having carboxyl group. Wood
Sci. Technol. 21 : 75-88.
Micales J.A., Hans J.S., David K.L. and Yong R.A. (1994). Fungi toxic activities of some pines cone
extractives. In Biotoxin, biodegradation and biodeterioration research. Ed. Plenum Press, G. C.
Lewellyn and C. E. O’Rear, New York.
Mitchell P.H. (1988). Irreversible property changes of small loblolly pine specimens heated in air,
nitrogen, or oxygen. Wood and Fiber Science, 20(3), 320–355.
Mori M., Aoyama M., Hokkaido S. (1997). Antifungal constituents in the bark of Magnolia obovata.
Holz als Roh und Werkstoff, 55(4), 275-278.
Morton L.H.G. and Eggins H.O.W. (1976). The effect of moisture content in wood on the surface
growth and penetration of fungi. Material und organismen, 11, 279-294.
Mouras S., Girard P., Rousset P., Permadi P., Dirol D. and Labat G. (2002). Propriétés physiques
de bois peu durables soumis à un traitement de pyrolyse ménagée. Annals of Forest
Sciences, 59, 317-326.
Mouzouras R. (1989). Soft rot decay of wood by marine microfungi. J. Inst. Wood Sci. II, 5, 196.
Neya B. (2002). Valorisation de la biomasse du Burkina Faso: Faisabilité de panneaux de particules
à base de déchets agricoles & recherche de nouveaux biocides pour la préservation du bois à
partir d’essences naturellement durables. Thèse Sciences du bois, Université Henri Poincaré,
Nancy 1.
113
Neya B., Mohamed H., Mathieu P., Gérardin P. (2004). On the durability of Burkea africana
heartwood: evidence of biocidal and hydrophobic properties responsible for durability.
Annals of Forest Science, 61(3), 277-282.
Nicholas D.D. (1973). Degradation and protection of wood. In "Wood deterioration and its
prevention by preservative treatments" Vol. 1. Syracuse University press. pp. 380.
Nicholas D.D. (1973). Wood deterioration and its prevention by preservative treatments Vol. 2.
Syracuse University Press. pp. 402.
Nunes L. and Dickinson D.J. (1996). The effect of boric acid on the protozoan numbers of the
subterranean termite, Reticulitermes lucifugus. International Research Group in Wood
Preservation. Doc. No. IRG/WP 96-10148. Stockholm, Sweden.
Nuopponen M., Vuorinen T., Jasma S., Viitaniemi P., (2004). Thermal modifications in softwood
studied by FT-IR and UV resonance Raman spectroscopies. J. Wood Chem. Technology.,
24(1), 13-26.
Nyamasyo G.H. (1995). Termites (Isoptera) are an indispensable component in the arthropod trophic
relationship in the Kenyan semi-arid ecosystem. Proceedings of the Second Regional
Workshop on termite research and control. Nairobi, Kenya. pp. 74.
Okario J. and Peden D. (1992). The growth performance of Grevillea robusta in the Highlands of
Uganda: In Grevillea robusta in Forestry and Agro Forestry (Ed). C.E. Harwood, ICRAF,
Nairobi, Kenya.
Okwara N.D. (2000). Effect of site factors on the performance of CCA treated E. saligna poles.
M.phil Thesis, Moi University, Kenya.
Pandey K.K. and Pitman A.J. (2004). Examination of lignin content in softwood and a hardwood
decayed by brown-rot fungi with acetyl bromide method and Fourier transform infrared
spectroscopy. Journal of Polymer Science, 42, 2340-2346.
Patzelt M., Stingl R., Teischinger A. (2002). Termische Modifikation von Holz und deren
Einfluβ auf ausgewählte Holzeigenschaften, In Modifiziertes Holz Eigenschaften und
Märkte, LIGNOVISIONEN Band 3, ISSN 1681-2808, 101-149.
Peralta R.C.G., Menezes B., Carvalho A.G. and Aguiar-Menezes E. (2004). Wood consumption
rates of forest species by subterranean termites (Isoptera) under field conditions, Sociedade
de Investigações Florestais, 28, 283–289.
Pétrissans M., Gérardin P., El Bakali I., Serraj M. (2003). Wettability of heat-treated wood.
Holzforschung, 57(3), 301-307.
Pizzi A., Conradie W.E and Bariska, M. (1986). Polyflavonoid tannins - from a cause of CCA soft
rot failure to the "missing link" between lignin and micro distribution theories. Int. Res. Group.
Wood Preservation. Doc. No.3359.
114
Plaisantin H, Pailler R, Guette A, Birot M, Pillot JP, Daude G, Olry P. Ex-cellulose carbon fibres
with improved mechanical properties. J Mater Sci 2006;41:1959-1964.
Presnell T.L. and Nicholas D.D. (1990). Evaluation of combinations of low hazard biocides in
controlling mold and stain fungi on southern pine. Forest Products Journal, 40(2), 57-61.
Preston A.F. (2000). Wood preservation: trends of today that will influence the industry tomorrow.
Forest Products Journal, 50, 13-19.
Reyes Chilpa R., Gomez-Garibay F., Moreno-Tores G., Jimenez-Estrada M., Quiroz-Vasquez R. I.
(1998). Flavanoids and isoflavanoids with antifungal properties from platymiscium yucatanum
heartwood. Holzforschung, 52(5), 459-462.
Ridley D.D., Ritchie E., Taylor W.C. (1968). Chemical studies of the Protaceae II. Constituents
of Grevillea robusta; the synthesis of (n-tridecylresorcinol (grevillol) and related substances,
Australian Journal of Chemistry, 21(12), 2979-2988.
Ritchie E., Taylor W.C., Vautin S., Protaceae I. (1965). Grevillea robusta and Orites excelsa,
Austalian Journal of Chemistry, 18(12), 2015-2020.
Rowell RM, Le Van-Green SL Thermal Properties. In: Handbook of Wood Chemistry and Wood
Composites. Taylor & Francis, 2005. p. 121-138.
Rudman P. (1963). The causes of natural durability in timber, XI: Some tests on toxicity of wood
extractives and related compounds. Holzforschung, 17, 54-57.
Sailer M., Rapp A. O., Leithoff H., Peek R. D. (2000). Upgrading of wood by application of an oil-
heat treatment. Bunderesforschungsanstalt fur Forst- und Holzwirtscharft (BFH). Holz als Roh-
und Werkstoff, 58(1-2), 15-22.
Santos J.A. (2000). Mechanical behaviour of eucalyptus wood modified by heat. Wood Sciences
Technology, 34, 39-43.
Schirp A. and Wolcott M.P. (2005). In fluence of fungal decay and moisture adsorption on
mechanical properties of extruded wood-plastic composites. Wood and Fiber Science, 37(4),
643-653.
Schultz T.P., Harms W.B., Fisher T.H., Mc Murtrey K.D., Minn J. and Nicholas D.D. (1995).
Durability of angiosperm heartwood: The importance of extractives. Holzforschung, 49(1), 29-
34.
Schultz T.P., Darrel D.N., and Fisher T.H., (1997). Quantitative structure activity relationships of
stilbenes and related derivatives against wood-destroying fungi. Recent Res. Deel: In
Agriculture Food Chem., 1, 289-299.
115
Schultz T.P., and Nicholas D.D. (2000). Naturally durable heartwood: Evidence for the proposed
dual defensive function of extractives. Phytochem., 54, 47-52.
Schultz T.P., Nicholas D.D., Henry W.P., Pittman C.U. Wipf D.O., Goodell B. (2005). Review of
laboratory and outdoor exposure efficacy results of organic biocide: Antioxidant combinations,
an initial economic analysis and discussion of a proposed mechanism. Society of Wood Science
and technology, 37(1), 175-184.
Scott G.A. (1968). Deterioration and preservation of timber in buildings. Longmans, London, UK.
Singh A.P. and Kim Y.S. (1997). Biodegradation of wood in wet environments: a review. Int.
Res. Group. Wood preservation. Doc. No. 97-10217.
Sivonen H., Maunu S.L., Sundholm F., Jämsä S., Viitaniemi P. (2002). Magnetic resonance
studies of thermally modified wood, Holzforschung, 56, 648-654.
Slabbert N. (1992). Complexation of condensed tannins with metal ions. In R.W. Hemingway and
P.E. Laks Editors, Plant Polyphenols, Plenum Press, New York pp 421-436.
Standard method of evaluating wood preservatives by field tests with stakes, (1993). American
Wood Preservers Association (AWPA) standard E7-93.
Standard method of evaluating wood preservatives in soil beds, (1994). American Wood
Preservers Association (AWPA) E14-94.
Standard laboratory method of evaluating resistance to subterranean termites, (1997). American
Wood Preservers Association (AWPA) standard E1-97.
Standard test method for insoluble lignin in wood. ASTM (1998) 04.10.D1106-96.
Stewart A.P., Ripley P.H., Hoegedal P., Klein U. (2003). Investigation of diverse reactions
following the use of Econor in pigs. Journal of Veterinary Pharmacology and Therapeutics,
26, 253.
Theodore, C.S. (1973). Microbiological degradation and the casual organisms In:
"Wooddeterioration and its prevention by preservation treatment." Nicholas, D. Ed). Vol. 1,
Syracuse University Press, New York. pp. 67.
Thévenon M.F., Roussel C. and Haluk J.P. (2001). Possible durability transfer from durable to
non-durable wood species. The study case of teak wood. The International Research Group
on Wood Preservation, 32nd Annual Meeting, Doc No. IRG/WP 01-10392, Nara, Japan.
Thieblot L.,Berthelay S. and Berthelay J. (1971). Action preventive et curative d’un extrait
d’écorce de plante africaine ‘Pygeum africanum’ sur L’adénome prostatique expérimental
chez le rat. Therapie XXVI, 575-580.
116
Thieblot L., Grizard G. and Boucher D. (1977). Etude du VI326 (Principe actif d’un extrait
d’écorce de plante africaine ‘Pygeum africanum’ sur l'axe hypophyso-génito surrénalien du
rat. Therapie, 32, 99-100.
Tjeerdsma B.F., Boonstra M., Pizzi A., Tekely P. and Militz H. (1998). Characterisation of the
thermally modified wood: molecular reasons for wood performance improvement. Holz
Roh-Werkstoff, 56, 149-153.
Toussaint L. (1997). Utiliser les tannins pour la protection du bois. Telex bois, 4, 12.
Unsal O. and Ayrilmis N. (2005). Variations in compression strength and surface roughness of
heat treated Turkish river gum (Eucalyptus camaldulensis) wood. Journal of Wood Sciences,
51, 405-409.
Venkatasamy R. (1997). The wood preservation industry in Kenya. IRG/WP/97-30157.
Venkatasamy R. (2000). Wood Preservation in Kenya - A Report. The International. Res. Group
on Wood Preservation. Doc. IRG/WP00-40190.
Viitanen H. and Ritschkoff A-C. (1991a). Mould growth in pine and spruce sapwood in relation
to air humidity and temperature. Report No. 221, Swedish University of Agricultural
Sciences, Department of Forest Products, Uppsala, Sweden.
Viitanen H. and Ritschkoff A-C. (1991b). Brown rot decay in wooden constructions. – Effect of
temperature, humidity and moisture. Report No. 222, Swedish University of Agricultural
Sciences, Department of Forest Products, Uppsala, Sweden.
Viitaniemi P. (1993). Wood modification using heat treatment. Espoo, Finland: Research Projects
of the Forest Products Laboratory.
Walter S. and Rakotonirina J.C.R., (1995). Review of exploitation de Prunus africanum: A
Madagascar rapport elabore pour le PCDI zahumena et la direction des eaux et forest by Ian
Dawson. J. Agro forestry Today, 9(2), 15-17.
Weiland J.J. and Guyonnet R. (2003). Study of chemical modifications and fungi degradation of
thermally modified wood using DRIFT spectroscopy. Holz Roh-Werkstoff, 61, 216-220.
Wikberg H.and Maunu S.L. (2004). Characterisation of thermally modified hard- and softwoods
by 13C CPMAS NMR. Carbohyd. Polym., 58, 461-456.
Wilkinson F.G. (1979). Industrial Timber Preservation. Associated Business Press, London.
Windeisen E., Wegener G., Lesnino G., Schumacher P. (2002). Investigation of the correlation
between extractives content and natural durability in 20 cultivated larch trees. Holz als Roh
und Werstoff, 60(5), 373-374.
117
Yildiz S., Gezer E.D., Yilditz, U.C. (2006). Mechanical and chemical behaviour of spruce wood
modified by heat. Building and Environment, 41, 1762-1766.
Zabel R.A. and Morrell J.J. (1992). Wood Microbiology: Decay and its prevention. Academic
Press, USA.
Zawadzki J. and Wisniewski M. (2002). 13C NMR study of cellulose thermal treatment, Anal.
Appl. Pyrolysis, 62, 11-121.
Appendix 1
Francis MBURU1,2
Fred MUlSU2
Peter SIRMAH 2
Philippe GERARDIN1
1 Laboratoire d'études et derecherche sur le matériau bois(Lermab)Équipe de chimie organiqueet microbiologieUniversité Henri Poincaré· Nancy 1BP 23954506 Vandœuvre-les-Nancy CedexFrance
2 Department of Wood Scienceand TechnologyMoi UniversityPO Box 1125EldoretKenya
"BOIS ET FORËTS DES TROPIQUES, 2005, N' 286 (4) 65IMPRÉGNABILITÉ 1 LE POINT SUR ...
Impregnability of Grevillearobusta using the sapdisplacement method
To improve the durability of Grevillea robusta timber and thereby promoteits use in the construction sector, we have studied its impregnability by sapdisplacement method using conventional copper, chromium and arsenic mixture.Recommendations are made to obtain a high degree of protection under differentconditions, including uses in contact with the soil or immersion in fresh or marine water.
Grevillea robusta trees planted in a hedge for harvesting as timber.Photo F. Mburu.
/66 BOIS ET FORËTS DES TROPIQUES, 2005, W 286 (4)
FOCUS 1 IMPREGNABILITY
RÉSUMÉ
IMPRÉGNABILlTÉ DE GREVILLEAROBUSTA PAR LA MÉTHODEDU DÉPLACEMENT DE SÈVE
(ABSTRACT
IMPREGNABILITY OF GREVILLEAROBUSTA USING THE SAPDISPLACEMENT METHOD
Francis MBURU, Fred MUI5U,Peter SIRMAH, Philippe GERARDIN
l RESUMENIMPREGNABILIDAD DE GREVILLEAROBUSTA POR EL MÉTODO DEDESPLAZAMIENTO DE LA SAVIA
L'utilisation du bois dans la construction peut être facilitée par le recours àdes traitements de préservation peucoûteux permettant de limiter les fraisde remplacement et de conserver lesressources naturelles. Le déplacement de sève est une méthode simpleà mettre en œuvre facilement utilisable en milieu rural. Une étude a étéréalisée pour évaluer l'efficacité decette technique afin de promouvoirl'utilisation d'une essence peudurable du Kenya, Grevillea robusta.Des échantillons préalablement écorcés ou non ont été traités avec unmélange de cuivre, chrome et arsenic,dilué à 2,4 ou 6 %. Les taux de rétention ont été mesurés et comparésavec les valeurs théoriques recommandées pour ce type de traitement.La pénétration et le taux de rétentiondu produit de préservation augmentent avec la quantité de produit depréservation utilisée. Les échantillonsécorcés présentent généralement unemeilleure imprégnabilité que leurshomologues avec écorces et les tauxde rétention sont, dans tous les cas,supérieurs aux valeurs recommandées par la Fao.
Mots·c1és : Grevillea robusta, déplacement de sève, mélange cuivrechrome-arsenic, rétention, pénétration.
The utility of construction timber canbe improved through law-cast treatment methods that reduce the cost ofreplacing structural timbers replacement and thus help ta conserveforests. The sap displacementmethod is simple and easily appliedin rural areas. A study was carried outta investigate the effectiveness ofthistechnique with Grevillea robusta witha view to promoting this non- durableKenyan species. Specimens with andwithout bark were treated with a mixture of copper, chromium and arsenicin concentrations of 2,4 and 6%.Retention was determined and compared with the recommended values.Retention and penetration of thepreservatives increased significantlywith increased concentrations. Retention in debarked specimens was significantly higher than in un-debarkedspecimens, but in bath cases, reten·tion was higher than the recommended FAO values for various uses.These results suggest that the sapdisplacement method can be usedlocally as an effective treatment forGrevillea robusta timber for end usessuch as fencing pales.
Keywords: Grevil/ea robusta, sap displacement, copper-chromium-arsenicmixture, retention, penetration.
El usa de la madera en la construccion puede verse facilitado medianteel empleo de tratamientos de proteccion asequibles que permiten limitarlos costos de reemplazo y conservarlos recursos naturales. El desplazamiento de savia es un método deaplicacion simple, facilmente utilizable en el media rural. Se realizo unestudio para evaluar la eficacia déeste procedimiento para fomentar elusa de una especie poco duradera deKenia, Grevillea robusta. Algunasmuestras, previamente descortezadas a no, se trataron con una mezclade cabre, crama y arsénico diluida al2, 4 a 6 %. Se midieron las tasas deretencion y se compararon con losvalores teoricos recomendados paraeste tipo de tratamiento. La penetracion y la tasa de retencion dei producto de proteccion aumentan con lacantidad de producto de proteccionutilizada. Las muestras descortezadas suelen tener una mejar impregnabilidad que cuando tienen cortezay las tasas de retencion son, en todoslos casos, superiores a los valoresrecomendados par la FAO.
Palabras clave: Grevillea robusta,desplazamiento de savia, mezcla decob re-crom a-a rsén i co, retencion,penetracion.
The context
Wood is widely used for construction because of its physicalcharacteristics and engineering properties, associated with ready availability (EATON, HALE, 1993; ASTON,1985). As the technology advances,more diverse and complex uses ofwood have been developed. Despiteits versatility as a construction material, modern technology has broughtalternatives, such as metal and plastics, some of whose qualities havesuperseded those of wood.
The natural durability oftimber isfairly often a disadvantage comparedto other competing construction materials. Durability is mainly dependenton the wood structure of individualspecies and the chemical compositionof extractives (FENGEL, WEGENER,1983). The heartwood is generallydescribed as more resistant than sapwood to fungi and insect attacks.
ln Kenya, state forest plantationscover about 1 405 000 ha, or 2.4% ofthe total land area. These forests aremade up of 31 % pine, 45% cypress,10% eucalyptus and 14% of otherspecies. Private forests cover about70 000 ha and consist mainly of acacia species and Grevi//ea robusta, thelatter accounting for about 75% of thetotal area (KFMP, 1994). Acacia is usedto extract tannin while Grevillea provides multipurpose timber from agroforestry. The use of Grevi//ea for fencing and construction purposes hasbeen necessitated bya timber deficitin Kenya that appeared in 2000 due toindustrial development and a ban onlogging in government forests in 2001.Increasing population pressure hascaused depletion of natural forests,leading to shortages in industrial rawmaterial. As a result ofthis and the current ban on logging in Kenya, consumers have sought alternativesources and species for manufacturingand construction, including Grevi//earobusta. Grevi//ea robusta is a fastgrowing agro-forestry tree species distributed in most parts of Kenya'sCentral and Rift Valley provinces (KFMP,1994). Most fast growing trees have a
high ratio of sapwood to heartwood,and therefore have lower natural durability than selected slow-growing treesextracted from natural forests.
Various preservatives are used inKenya to suit different end uses. Theseinclude chromated copper arsenate(CCA) , pentachlorophenol (PCP) andcreosote oil. Pressure treatment ofEucalyptus saligna using tar oil ismainly used to achieve high rates ofretention (160-400 kg/m 3) as recommended for high hazard areas againsttermites and in marine waters.Eucalyptus saligna transmission poleshave also been treated with CCA usingthe full cell pressure method. However,premature failures have been reportedfor some poles in service due to lowpreservative fixation associated withthe anatomical characteristics of eucalyptus (OKWARA, 2000; VENKATASAMY,1997). Cheap preservation treatmentsincluding brushing, immersion or sapdisplacement can be investigated toimprove the durability of Grevilleawood. The effectiveness of these treatments depends not only on the methodused but also on the preservativeemployed. It is generally accepted thatsap displacement enables sapwood tobe treated effectively by allowing uniform penetration of the active ingredients used for preservation. This is acheap method that can be used to treatGrevi//ea robusta for local constructionuses in Kenya. The choice of preservative depends on the level of risk of bio-
Tree G. robusta pales used for agate.Photo F. Mburu.
"BOIS ET FORËTS DES TROPIQUES, 2005, W 286 (4) 67IMPRÉGNABILlTÉ 1 LE POINT SUR ...
logical attack on the wood during itsservice life. Because of the extent ofdecay and termite degradation, CCAwas retained as the most effective lowcost preservative to improve the durability and performance of Grevillearobusta as construction timber.
The sap-displacement methodmay be applied to freshly felled greenhardwood or softwood and uses theprinciple of hydrostatic pressure toforce the preservative from the butt endof the round timber to the top. Someimpermeable species like Picea spp.can take a long time to season adequately. Fowlie and Sheard reportedsuccessful treatment of Picea obies,Picea sitchensis and Pinus nigra in 2030 hours, using pressurized sap displacement (FOWLlE, SHEARD, 1983). Themethod produces a greater concentration of the preservative at the butt endand this is welcome, especially in thecase of poles driven into the ground,which are more vulnerable to decay.This method is also economicallyadvantageous in remote areas with nofacilities for timber impregnation andwhere labour costs are not too high(FINDLEY, 1985; GOODELL et al., 1991).
The aim ofthis study is to investigate the impregnability of Grevi//earobusta using sap-displacement toenhance its uses, reduce replacement costs and provide a relativelydurable construction material, especially in rural areas where it is mostlygrown in agro-forestry systems.
/68 BOIS ET FORËTS DES TROPIQUES. 2005. N" 286 (4)
FOCUS 1 IMPREGNABILITY
4540
Debarked specimens -----~M-35E Y =5.25 x + 8.003 .........
0> 30 R2 =0.915 --~c 25 .'0
20 --~Q) 15 Undebarked specimens'li>a: 10 Y =3.122 x + 5.573
5R2 =0.960
0
0 2 3 4 5 6 7
CCA solution concentration (%)
Figure 1.Relationships between retention and concentrations of chromated copperarsenate (CCA) solutions.
Two Grevi/lea robusta trees.Photo F. Mburu.
Materialsand methods
Sampling
Specimens from freshly cut lmlengths of roundwood from twentyyear-old Grevillea robusta A. Cunn ex.R. Br. with a top diameter of 140180 mm were cut following the procedure of HARWOOD and BOOTH (1992).SixtYspecimens were prepared: thirtywere debarked and thirty undebarkedbut with a small debarked portion toallow preservative uptake. Thedebarked portion is dipped into theCCA container to avoid direct absorption of the chemical by the bark. Theuptake of CCA from the container wasexpected to be mainly by suction created through sap displacement.
Sap-displacementtreatment
Industrial-quality CCA was usedduring this study (Tanalith-C, ArchTimber). Aqueous solutions of CCAwere prepared at different concentrations (2. 4 or 6% in mass) and usedto impregnate the specimens. Thebutt end of each specimen wasplaced in a plastic container fi liedwith CCA solution (five litres fordebarked and four litres for undebarked poles). The container wasfirmly attached to the butt end andsupported in a slanting position. Thetreatment was continued until thechemical oozed out from the top endof the posts or when ail the preservative in the container was absorbed.At the end of the treatment the containers were removed and the specimens were left in the same positionfor 7 days to leach out unfixed preservative. Retention (kg/m)) wasdetermined for both debarked andun-debarked specimens and compared with the recommended average retention for timber under various exposure conditions according tothe formula (i):
"BOIS ET FORÊTS DES TROPIQUES, 2005, N° 286 (4) 69IMPIlÉGNABILlTÉ 1 LE POINT SUR ...
Table 1.Mean chromated copper arsenate (CCA) retention values measured and possibleuses according to FAO standards.
Timber in contactwith the ground
Permanently immersedin fresh water
Permanently immersedin sea water
25.04 ± 3.26
12.55 ± 1.73
16.60 ± 1.68
Undebarked specimens
CCA Retention Possible use(kg/m 3) according to FAO
16.66 ± 1.70
37.66 ± 5.60*
32.69 ± 2.97*
Debarked specimens
CCA Retention Possible use t(kg/m3) according to FAO
Permanently immersedin fresh water
Permanently immersedin sea water
Permanently immersedin sea water
* Significantly different values at 5% level of significance.
2
6
4
Concentrationof treatmentsolution (%)
ImpregnabilityRetention = (m 2-m l ) xc / v ... (i)
Retentionwhere ml is the mass of the
fresh untreated sample (kg), m2 isthe weight of the impregnated sampie measured immediately after theCCA solution had oozed out from thetop of the pole (kg), c is the CCA solution concentration (%) and v the sampie volume (m 3).
Penetration was measured inpercentage along the height of sampies according to the formula (ii):
Penetration = (l-[r·r1J/r) x 100 ... (ii)
where r1 is the depth of preservative penetration (cm) and r theradius of the sam pie (cm) for a givenposition from the butt end.
The Food and AgricultureOrganisation (FAO) , recommends aCCA retenti on value of 6 kg/m 3 forinterior timbers not in contact withthe ground, such as trusses or rafters,and of 8 kg/m 3 for exterior timbersnot in contact with the ground, suchas doors and windows. The recommended value for timber in contactwith the ground, such as fence posts,railway sleepers or bridge timbers, is12 kg/m 3 • Timber permanentlyimmersed in fresh water requires aretention value of 16 kg/m3 and timber immersed in seawater (e.g.groynes, jetties or boat building timber) requires 24 kg/m 3 (FAO, 1986).
Table 1 shows mean CCA retention values obtained for debarkedand undebarked specimens and theirpossible uses according to FAO recommendations.
Debarked specimens had significantly higher retention than theundebarked specimens for the samesolution strength. In both debarkedand undebarked specimens, retention increased with the strength ofthe solution.
Comparisons between theexperimental mean retention valuesand the FAO recommended valuesindicated that the impregnation of G.robusta using the sap displacementmethod is sufficient to reach thevalue of 12 kg/m 3 required for usesin contact with the ground. For the2% solution, debarked specimensdid not have a significantly differentvalue from the value recommendedfor immersion in fresh water. Thisimplil2s that G. robusta meets theminimum requirements set for timbers to be permanently immersed infresh water. However, the valuesobtained with the 4 and 6% solutionswere significantly higherthan the recommended values. These are suitable for ail environmental hazardsbut they are uneconomical becauseof the high retention obtained.
G. robusta posts and runners forming a fence,Photo F. Mburu.
/'70 BOIS ET FORÊTS DES TROPIQUES, 2005, N° 2B6 (4)
FOCUS 1 IMPREGNABILITY
CCA solution concentration (%)
Table Il shows penetration indebarked and undebarked specimens using 2, 4 and 6% solutions.
With ail three concentrations,penetration was higher in debarkedspecimens.
Retention = 5.25 (solutionstrength) + 8.00 (iii)
Retention = 3.12 (solutionstrength) + 5.57 (iv)
Values obtained for undebarked specimens using the 2, 4and 6% solutions were generallysmaller than those obtained fordebarked specimens. This may bebecause the larger surface areaincreased suction pressure andtherefore absorption of the chemical.However, these values are in ail caseshigher than the 12 kg/m 3 recommended by FAO. G. robusta treatedwith 2 and 4% solutions of CCA canmeet the minimum retention requirements for uses in contact with theground and immersion in fresh water,respectively. Values obtained for the6 % CCA solution were slightly higherand allow uses where timbers arepermanently immersed in sea water.
Figure 1 shows the relationshipbetween retention and solutionstrength for debarked and undebarked samples. The linear relationship is described by equations (iii)and (iv) and the corresponding correlation coefficients of R2 = 0.92 andR2 = 0.96 respectively:
Retention in undebarked specimens has a stronger linear relationship than in debarked specimens.These models can be used to determine levels of retention in bothdebarked and undebarked specimens using the sap displacementmethod.
The higher retention valuesmeasured in debarked samples maybe due to the larger surface area creating increased suction pressures,which cause evaporation and higherchemical uptake.
Penetration
8
1000
• Treated at 2%
• Treated at 4%
t. Treated at 6%
1000
t. Treated at 6%
• Treated at 2%
• Treated at 4%
___ -11
800
6
Undebarked specimensy =2,365 x + 59.837
R2 =0.904
800
4
600
--_11-----------
400
Height (mm)
2
fl- -
200 400 600
Height (mm)
200
Debarked specimensy = 2.204 x + 76.327
R2 =0.996
Figure 2.Relationships between depth of penetration and chromated copper arsenate(CCA) concentration measured at a height of 500 mm.
Figure 3.Chromated copper arsenate (CCA) penetration according to extraction heightfor debarked specimens.
100
90
'$. 80c 700
~ 60Q)c 50Q)
0..40
300
100
'$.90
c 800
~ 70Q)cQ)
600..
50
0
95
90
85
~80c
0
~ 75Q)cQ) 700..
65
60
55
0
Figure 4.Chromated copper arsenate (CCA) penetration according to extraction heightfor undebarked specimens.
"BOIS ET FORÊTS DES TROPIQUES, 2005, N' 286 (4) 71
IMPRÉGNABILlTÉ 1 LE POINT SUR ...
Table Il.Mean penetration for debarked and undebarked specimens.
Penetration (cm)
Debarked specimens Undebarked specimens
Transverse section of a treated G. robusta pole.Photo F. Mburu.
63.7 ± 21.7
71.1±17.6
73.2 ± 6.9
with the height of the specimen. Thedepth of penetration was alsodependent on the concentration ofthe treatment solution, i.e. thestronger the concentration, thehigher the penetration and consequently the retention.
80.6 ± 12.7
85.6 ± 9.8
89.4 ± 7.8
Concentrationof CCA solution (%)
2
4
6
There were significant differencesin penetration between debarked andundebarked specimens. Preservativepenetration increased with higher solution strengths. Debarked specimensshowed higher penetration than undebarked specimens. Debarked sampleshave a higher surface area for evaporation, which creates a higher suctionpressure than in barked samples,hence higher penetration. Even intreatments using the same solutionstrength, samples showed variation inpenetration. This can be explained bydifferences in sapwood content, density and moisture content. There is astronger linear relationship betweenpenetration and solution strength fordebarked than for undebarked specimens. The strong correlation betweenpenetration and solution strength forthe two groups of specimens indicatesthat the two models adequately predictthe degree of preservative penetration.
Variations in the depth of penetration of the preservative accordingto the height of specimens are shownin Figures 3 and a 4 for debarked andundebarked samples respectively. Inboth cases, penetration decreased
Penetration = 2.21 (solutionstrength) + 76.3 (v)
Penetration = 2.37 (solutionstrength) + 59.8 (vi)
Figure 2 shows the variation inpreservative penetration according tosolution strength for debarked andundebarked samples. The linear rela·tionship is described by equations(v) and (vi) and by the correspondingcorrelation coefficients of R2 = 0.97and R2 = 0.90 respectively:
Mature G. robusta trees showing orange flowers.Photo F. Mburu.
/72 BOIS ET FORÊTS DES TROPIQUES, 2005, W 286 (4)
FOCUS 1 IMPREGNABILITY
Mature 24 year-old G. robustaplantation, growing in the UasinGishu district of the Rift valleyProvince. Heavy rainfall area, highaltitude plain.Photo F. Mburu.
Conclusion
Our results show that Grevi/learobusta can be adequately treatedusing the sap displacement methodto retention levels that meet minimum requirements for various useconditions, including contact withthe ground or permanent immersionin fresh and sea water. The positivelinear relationship between retention, penetration and solutionstrength means that the desiredretention and penetration of preservative can be achieved by varying thesolution strength. Preservative penetration is dependent on the height ofthe pole and inversely proportional tothe distance from the end dipped inpreservative. Undebarked specimenstend to retain less preservative, whiledebarked ones yield greater penetration.
Debarked G. robusta poles canachieve the degree of retentionrequired for wood permanentlyimmersed in fresh water with a 2%solution and for marine water exposure with 4 and 6% solutions.Undebarked wood can be treated toachieve the retention levels requiredfor ground contact using a 2% solution, and for fresh water and marineexposure using 4 and 6% solutions.
G. robusta trees ready forharvesting and use as fence posts.Photo F. Mburu.
References
ASTON O., 1985. CCA preservativesand their application to timbers in thetropics. In: Preservation of timber inthe tropics. Dordrecht, Netherlands,Martinus Nijhoff and Dr. Junk publishers, FindleyW.P.K.(ed.), 141-155.
EATON RA, HALE M.D.C., 1993.Wood: decay, pests and protection.London, United Kingdom, Chapman& Hall, 99 p.
FAO, 1986. Wood preservation manual.Rome, Italy, Food and Agriculture Organization of the United Nations, 151 p.
FENGEL O., WEGENER G., 1984. Woodchemistry, ultrastructure, reactions.Berlin, Germany, Walter de Gruyter &Co, 183-226.
FINDLEY W.P.K., 1985. The nature anddurability of wood, Preservation oftimber in tropics. Dordrecht, Nethetlands, Martinus Nijhoff and Dr. Junkpublishers, Findley W.P.K.(ed.), 1-13.
FOWLIE I.M., SHEARD L., 1983. Developments in the use of home grownspruce poles for use as overhead.Record of the Annual Convention ofthe British Wood Preserving Association, 47-60.
GOODELL B.F., KAMKE A., LIU J., 1991.Laser incising of spruce lumber forimproved preservative penetration.International Research Group on WoodPreservation, IRG/WP/91-3646.
HARWOOD C.E., BOOTH T.H., 1992.Status of Grevillea robusta in Forestryand Agroforestry. Nairobi, Kenya, CE.Harwood (ed.), ICRAF, 9-16.
KFMP, 1994. Kenya forest master plandevelopment programmes. Ministry ofEnvironment and Natural Resources,Government of Kenya, 1-208.
OKWARA D. N., 2000. Effect of sitefactors on the performance of CCAtreated Eucalyptus saligna poles.Thesis, Moi University, Kenya, 92 p.
VENKATASAMY R.N., 1997. The woodpreservation industry in Kenya. International Research Group on WoodPreservation,IRG/WP/97-30157.
Appendix 2
Evaluation of thermally modified Grevillea robusta heartwoodas an alternative to shortage of wood resource in Kenya:
Characterisation of physicochemical properties and improvementof bio-resistance
Francis Mburu a,b, Stephane Dumarcay a, Francoise Huber a, Mathieu Petrissans a,Philippe Gerardin a,*
a Laboratoire d’Etudes et de Recherches sur le Materiau Bois, UMR INRA 1093, Universite Henri Poincare Nancy 1, Faculte des Sciences et Techniques,
BP 239, 54506 Vandoeuvre-les-Nancy, Franceb Department of Wood Science and Technology, Moi University, P.O. Box 1125, Eldoret, Kenya
Received 12 September 2006; received in revised form 27 October 2006; accepted 4 November 2006Available online 29 December 2006
Abstract
Heat treatment of Grevillea robusta, a tropical wood species of low natural durability, was carried-out under inert conditions toimprove its decay resistance. Resistance of heat treated samples was evaluated by malt agar block tests after three months of exposureto several wood rotting fungi. Also resistance of heat treated wood against termites was tested in the laboratory and in the field. Resultsshowed that durability against fungi and termites was greatly improved after treatment. There was a good correlation between decayresistance and mass loss due to thermal treatment. Microscopic, FTIR and 13C MAS NMR analysis were performed to characterizewood chemical and anatomical modifications that occur after treatment to understand the reasons of the durability improvement.� 2006 Elsevier Ltd. All rights reserved.
Keywords: Durability; Grevillea robusta; Heat treatment; Rotting fungi; Termite
1. Introduction
In Kenya, forests resources rank high among otherimportant natural resources. Forest ecosystems are com-plex natural resource base which provides environmentalgoods and services for social, cultural and economic devel-opment. It is therefore important that the resource isconserved, protected and sustainably utilized for nationaldevelopment. Kenya like other developing countries isconfronted with challenges of over reliance on naturalresources.
Timber demand in Kenya is rising each year unlike itssupply. If this trend continues it is projected that by theyear 2020 Kenya will have a deficit of 6,841,000 m3 ofwood (KFMP, 1994). The deficit is increased by prematurefailures of wood in service as a result of biological deterio-ration, failures in establishing new and expanding forestplantations, repeated excisions with no compensatory addi-tions and changes of forest land use. One way of counter-acting this problem of timber deficit and deforestation is tomake the timber last longer in service, through effectivetreatment methods.
The Kenyan forest plantations are constituted of 31%Pines, 45% Cypress, 10% Eucalypts and others 14%, whichare the main base for industrial raw materials before thelogging ban effected by the government in the year 1999.Private forests cover about 70,000 ha consisting mainly
0960-8524/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2006.11.006
* Corresponding author. Tel.: +33 3 83 68 48 40; fax: +33 3 83 68 48 51.E-mail address: [email protected] (P. Gerar-
din).
Bioresource Technology 98 (2007) 3478–3486
acacia species and Grevillea robusta out of which G. robusta
is about 75% of the area. It is a fast growing multipurposeagro-forest tree, widely distributed in Kenya especiallyin the agricultural prime areas. Due to the logging banin the government forests, there has been acute shortageof wood for construction, fencing and other purposes.Increase of population in Kenya has also caused depletionof the natural forests leading to shortage in industrial rawmaterial. This has forced the farmers to look for alternativesources of wood. Most of them have resorted to useG. robusta due to its availability, but its low durabilityagainst bio-degraders is a great challenge to all wood usersand the Kenyan government. Presently, copper chromearsenate (CCA), pentachlorophenol (PCP) and creosoteoil are the most used biocides for wood preservation inKenya using dipping or pressure processes.
The increasing environmental pressure and awarenessover the last few years has led to an important change inthe field of wood preservation in industrialised countriesin regard to biocide toxicity. This has lead to developmentof ‘‘non biocidal’’ alternatives like chemical or thermaltreatments (Hill, 2005). Heat treatment has been particu-larly developed in Europe during this last decade leadingto industrialisation and commercialization of heat treatedtimbers resulting from the treatment of low natural dura-bility wood species like pine, spruce, poplar or beech (Pat-zelt et al., 2002; Alen et al., 2002). The aim of our study wasto investigate effect of such treatments on durability of G.
robusta, a tropical wood species of low natural durability,to allow its development and uses in Kenya for applica-tions where mechanical properties are not required. Indeed,it is known that an important drawback of heat treatedwood is mechanical brittleness (Santos, 2000; Mouraset al., 2002; Unsal and Ayrilmis, 2005; Yildiz et al.,2006). For these reasons, heat treated wood is not recom-mended for use in load-bearing constructions, but can findvaluable applications for furnitures, wall, ceiling, roofing,flooring and fencing.
2. Experimental
2.1. Material
G. robusta heartwood was used in this study. Woodblocks measuring 50 mm · 30 mm · 20 mm in longitudinal,radial and tangential directions were cut and oven driedat 103 �C (approximately 48 h) before determination oftheir anhydrous weights (m0).
2.2. Heat treatment
Heat treatment was performed on previously driedblocks in a reactor placed in an oven at different tempera-tures (240, 250 and 260 �C) during different time (0.5,1,5,7and 15 h) under a nitrogen atmosphere. The oven temper-ature was increased by 20 �C mn�1 from ambient to theoperating temperature. After the treatment, the tempera-
ture decreases slowly to the room temperature. Mass lossafter heat treatment was calculated according to theformula:
ML ð%Þ ¼ 100� ðm0 � m1Þ=m0;
where m0 is the initial oven dried mass of wood sample andm1 the oven dried mass of the same sample after heattreatment.
2.3. Exposure to fungi
Two kinds of experiments were performed to evaluatefungal durability after heat treatment. The first was resis-tance evaluation of samples treated at 250 �C for differentduration of time against Coriolus versicolor. The secondwas evaluation of heat treated blocks at 250 �C for 7 hagainst different wood rotting fungal species, two whiterots (C. versicolor and Pycnoporus sanguineus) and twobrown rots (Poria placenta and Antrodia sp.). The choiceof the heat treatment conditions (250 �C for 7 h) was basedon previously reported experiments performed on beechindicating that improvement of durability can be achievedfor mass losses of approximately 20% (Hakkou et al.,2006). Heat treated G. robusta blocks were cut into smallerblocks of 25 mm · 25 mm · 5 mm in longitudinal, radialand tangential directions. These were used for fungal dura-bility evaluation after conditioning in an oven at 103 �C forone night (m2). Petri dishes (9 cm diameter) were filled withsterile culture medium prepared by mixing 30 g malt and40 g agar in distilled water (1 l), inoculated with the differ-ent fungi and incubated at 22 �C and 70% relative humidityto allow full colonization by the mycelium. Two blocks(treated or untreated as control) were placed in each Petridishes and incubated during three months to evaluate theeffect of heat treatment. Each experiment was duplicatedor triplicated (see after). After this period, mycelia wereremoved and the blocks were dried at 103 �C and weighed(m3) to determine the weight loss caused by the fungalattack
WL ð%Þ ¼ 100� ðm2 � m3Þ=m2;
where m2 is the initial oven dried mass of wood blockbefore attack and m3 is the oven dried mass after attack.
Moisture content of the wood after fungal attack wasdetermined as followed:
MC ð%Þ ¼ 100� ðm30 � m3Þ=m3;
where m30 is the wet weight of the sample after attack mea-sured after removal of the mycelium and m3 is the ovendried mass after attack.
2.4. Exposure to termites
Heat treated G. robusta samples (7 h at 250 �C) werealso exposed to Macrotermes natalensis widely distributedin Kenya in laboratory conditions according to a proce-dure adapted from AWPA E1-97 standard (Standard
F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486 3479
method laboratory for evaluation to determine resistanceto subterranean termites, 1997). Pinus sylvestris anduntreated G. robusta were used as control. The tests weredone in chambers free of organic material in glass jars80 mm diameter by 100 mm in height. Prior to use, all con-tainers were sterilized in autoclave. One hundred and fiftygrams of sand were added to each container, followed by30 ml of distilled water then allowed to stand for 2 h.Two dried weighed (m2) test blocks, 30 mm · 10 mm ·20 mm in longitudinal, radial and tangential directions,were placed on the surface of each jar with two cornersagainst the side of the container. Four-hundred termiteswere added to each container at a ratio of 360:40 workersto soldiers respectively. All containers were maintained at25 �C for 28 days. Percentage change in dry mass of thetest block was determined.
Blocks were then dried at 103 �C and weighed (m4) todetermine the weight loss caused by the termite attack
WL ð%Þ ¼ 100� ðm2 � m4Þ=m2;
where m2 is the initial oven dried mass of wood blockbefore attack and m4 is the oven dried mass after attack.
Resistance to termite was also investigated using fieldtest conditions. Tests were carried out according to amethod adapted from AWPA E7-93 standard (Standardmethod of evaluating wood preservatives by field tests withstakes, 1993) especially as concerns the sample size whichwas reduced to fit with our heat treatment equipment. Anactive termite nest constituted of M. natalensis was identi-fied and cleaned from all the cellulolytic materials. The sitewas then watered and covered with an opaque plastic sheetto activate the termites. Samples measuring 50 mm ·25 mm · 25 mm in longitudinal, radial and tangentialdirections respectively were buried down at a radius of1 m from the termite nest. Three kinds of samples wereused: 24 samples of heat treated G. robusta (7 h at250 �C), 24 samples of untreated G. robusta and 24 samplesof P. sylvestris. The moisture at the test site was improvedby watering to maintain and attract more termites. Evalu-ation was done every month by removing six stakes foreach treatment and the values averaged.
2.5. Determination of the amount of extractive
Untreated and heat treated G. robusta blocks wereground to sawdust and dried at 103 �C before extractionwith dichloromethane in a Soxhlet extractor for 16 h atthe rate of 8–12 cycles per hour. After extraction thesolvent was evaporated under reduced pressure and theresidue dried over P2O5 in a desiccator before weighing.
2.6. Spectroscopic analysis
2.6.1. FTIR analysis
FTIR spectra were recorded as KBr disks on a PerkinElmer FTIR spectrometer SPECTRUM 2000 betweenwave number range of 4000 and 400 cm�1. Finely divided
9 mg wood samples, obtained after grinding of blocksand drying at 103 �C, were dispersed in a matrix of KBr(300 mg) and pressed to form pellets. Due to the difficultyto define a reference spectral band, which remains com-pletely invariable, to perform quantitative measurements,all comparisons were made qualitatively.
2.6.2. CP/MAS 13C NMR analysis
Solid state CP/MAS (cross-polarisation/magic anglespinning) 13C NMR spectra were recorded on a BrukerMSL 300 spectrometer at a frequency of 75.47 MHz.Acquisition time is of 0.026 s with number of transientsof about 1200. All the spectra were run with a relaxationdelay of 5 s, CP time of 1 ms and spectral width of20,000 Hz. Spinning rates are 5 KHz. Chemical shifts areexpressed in parts per million (ppm).
2.6.3. 1H NMR analysis1H NMR spectra were recorded in CDCl3 on a Bruker
AM 400 spectrometer. Chemical shifts were expressed inppm and calculated relative to TMS.
2.7. Contact angles measurements
Contact angles of untreated and heat treated blockswere recorded with a goniometer using water as probeliquid. Five microliters of distilled water were depositedon the wood surface (tangential face) with a microsyringueand contact angle measured after the total relaxation of thedrop (equilibrium state). Six measurements were performedon each block and the values were averaged.
2.8. Microscopic analysis
Microscopic observation were performed with an envi-ronmental scanning electron microscope (ESEM Quanta200) on a small block of 20 · 15 · 15 mm (L,T,R). Thetransversal section of the block was prepared with a micro-tome, repaired with marks and observed under the scan-ning microscope. After observation, the sample was heattreated for 4 h at 250 �C under nitrogen and then, observedagain using scanning microscope at the marked points tocompare wood anatomy before and after treatment. Thismethodology allows to avoid problems of surface prepara-tion on heat treated samples which were generally verybrittle.
3. Results and discussion
Fig. 1 represents the evolution of mass loss according totime at different temperatures.
Degradation of G. robusta wood is important during thefirst stage of the thermal treatment process and becomesless important in a second stage. The mass loss increaseswith time and temperature. As reported in the literatureon European wood species (Nuopponen et al., 2004; Wik-berg and Maunu, 2004; Hakkou et al., 2005), these mass
3480 F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486
losses are probably due to important degradations ofhemicelluloses during the treatment. To confirm theseassumptions, FTIR and CP/MAS 13C NMR analysis wereperformed to characterize wood chemical components(Figs. 2 and 3).
CP/MAS 13C NMR spectra of untreated and thermallymodified G. robusta wood are presented in Fig. 2.Untreated sample indicates classical signals ascribable tomain wood polymer components. Cellulose appears inthe region between 60 and 105 ppm. The signals at 72–75 ppm are assigned to C-2,3,5 carbon, the signal at65 ppm assigned to C-4 carbon and this at 105 ppm to C-1 hemiacetallic carbon. C-6 signal of cellulose is duplicateddue to the presence of amorphous and crystalline celluloseappearing at 84 and 89 ppm respectively. Signals of hemi-celluloses are less obvious due to their overlapping withthose of cellulose. The shoulder at 102 ppm on the C-1 sig-nal of cellulose is assigned to hemiacetallic carbon of hemi-celluloses. Acetyl groups of hemicelluloses are detected at20 and 173 ppm. Methoxyl groups of syringyl and guaiacylunits of lignin appears at 56 ppm, while aromatic carbonsappear between 120 and 160 ppm. Aliphatic carbons ofphenylpropane units of lignin are partially masked by poly-saccharidic signals.
Evolution of NMR spectra recorded after heat treatmentindicates that the main modifications appear rapidly afterthe first hour of treatment. Degradation of hemicellulosesis particularly obvious. The signal of the methyl of the ace-tyl groups at 20 ppm, mainly present on hemicelluloses,disappears totally after heat treatment. The shoulder at102 ppm on the C-1 signal of cellulose at 105 ppm decreasesas well after heat treatment confirming hemicelluloses deg-radation. Similar observations have been reported in theliterature on different non durable European species(Tjeerdsma et al., 1998; Sivonen et al., 2002; Weiland andGuyonnet, 2003). As reported in the literature, cellulosecrystallinity was more important after heat treatment. Thesignal at 89 ppm due to C-4 of crystalline cellulose increasesslightly compared to the signal at 82 ppm due to amorphouscellulose C-4 (Wikberg and Maunu, 2004). General behav-iour of NMR spectra is similar to spectra of other woodspecies reported in the literature excepted for the signal at30 ppm which is very unusual. Explanation of this signal
is due to the presence of important quantity of lipophilicextractives, approximately 5%, identified after Soxhletextraction with dichloromethane and 1H NMR analysis tobe mainly constituted of n-alkyl and n-alkenylresorcinols(Ritchie et al., 1965; Ridley et al., 1968; Cannon et al.,1973). Changes in the lignin signals are less significant indi-cating its higher stability. The appearance of new broadaromatic or alkenic resonances (110–135 ppm) and ali-phatic resonances (10–50 ppm) after heat treatment indicatestructural modifications of wood polymers due probablyto a beginning of destruction of the amorphous cellulose
0
10
20
30
40
50
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Time (h)
ML
(%)
260 °C
250 °C
240 °C
Fig. 1. Mass loss of Grevillea robusta wood after heat treatment.
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Abs
olut
e In
tens
ity
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Nor
mal
ized
Inte
nsity
220 200 180 160 140 120 100 80 60 40 20 0 -20 -40Chemical Shift (ppm)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Nor
mal
ized
Inte
nsity
Heat treated at 250 ˚C during 1 hour
Heat treated at 250 ˚C
during 7 hour
72
Control
20
30
56
65
75
84
105
89
175 15
4
135
Fig. 2. CP/MAS 13C NMR spectra of Grevillea robusta wood afterthermal treatment.
F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486 3481
structure. Such modifications, previously reported duringcellulose thermal treatment (Zawadzki and Wisniewski,2002), could be at the origin of the increase of crystallinity.
FTIR analysis are more difficult to interpret due to thesimilarities of the spectra before and after treatments.However some small modifications can be observed indi-cating the same tendencies as previously observed byNMR. The O–H absorption band at 3350 cm�1 decreasedcompared to all other bands. Characteristic C–O absorp-tion band of polysaccharidic components at 1058 cm�1
corresponding to C–O stretching vibrations decreasesslightly comparatively to lignin characteristic band at1600, 1510 and 1450 cm�1 confirming the degradation ofhemicelluloses.
Microscopic analysis of G. robusta wood before andafter heat treatment is presented in Fig. 4.
Microscopic analysis indicates that anatomical structureof wood was only slightly affected during treatment(Fig. 4). Vessels, fibers, parenchyma and rays are still obvi-ous after heat treatment. The main differences was presence
4000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.010.00
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0
65.00
cm-1
%T
4000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.021.53
25.0
30.0
35.0
40.0
45.0
50.0
54.60
cm-1
%T
4000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.02.29
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
47.40
cm-1
%T
Heat treated at 250˚Cduring 1 hour
Control
Heat treated at 250 ˚Cduring 7 hours
3350
1600
1510
14
50
1058
3350
1600
15
10
1450
1058
3350
1600
15
10
1450
1058
Fig. 3. FTIR spectra of Grevillea robusta wood after thermal treatment.
3482 F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486
Fig. 4. Transverse sections of Grevillea robusta wood.
ppm (f1) 1.02.03.04.05.06.07.0
7,28
6,26
6,19
5,37
2,50
2,04
1,59
1,28
0,91
ppm (f1) 1.02.03.04.05.06.07.0
7,19
3
3,89
0
1,19
1
A
B
R
HO
HO
R = -(CH2)nCH3 or -(CH2)nCH=CH(CH2)n' CH3
Fig. 5. 1HNMR spectra of dichloromethane extractives of Grevillea robusta wood before (A) and after thermal treatment at 250 �C during 7 h (B).
F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486 3483
of important quantities of extractives deposited in the ves-sels, which disappear after thermal treatment. 1H NMRanalysis of extractives of untreated and heat treatedG. robusta after Soxhlet extraction with dichloromethaneindicated an important degradation of these latter ones(Fig. 5). Before heat treatment, extracts are mainly consti-tuted of n-alkyl and alkenyl resorcinols. Chemical shifts ofaromatic protons appears as two singlets at 6.19 and6.26 ppm characteristic respectively of the proton betweenthe two hydroxyl groups and of the two other protons. Sig-nal at 5.37 ppm is attributed to vinylic protons of ethylenicdouble bond. Signals at 2.50 and 2.04 ppm are characteris-tic of benzylic and allylic protons, while signals at 1.59, 1.28and 0.91 correspond to methylene and methyl groups offatty alkyl chain. Proton of phenolic hydroxyl groupappears as a singlet at 7.28 ppm. 1H NMR spectra isstrongly modified after heat treatment indicating a totaldisappearance of characteristic chemical shifts of extractsinitially present in wood. Signals of aromatic and ethylenichydrogen atoms totally disappeared, while those of fattyalkyl chain are still obvious but present important modifi-cations. The assignment of the NMR signals of extractivesafter heat treatment are not clearly identified at present. Inthe same time, quantity of extractives decreased from 5.2%before heat treatment to 1.6% after heat treatment confirm-ing the degradation observed by NMR. Heat treatmentresult also in some shrinkage. Due to the anisotropy ofwood, the shrinkage depends on the orientation. Tangen-tial shrinkage is of approximately 6%, while radial shrink-age is of approximately 2% as measured with a veneercalliper.
Effect of time of heat treatment performed at 250 �C onthe durability to the white rot fungus C. versicolor isreported in Table 1. The results shows that heat treatmentconsiderably improves wood durability. Different levels ofwood degradation corresponds to different weight lossesafter varying the treatment time. It was observed that theability of C. vericolor to degrade wood was reduced aftermodification by heat treatment. Durability against fungusincreased with treatment time with optimum resistancereached after 20% mass losses. At the same time, untreatedG. robusta control and beech wood control virulence were
severely degraded by the fungus showing validity of theexperiment. Decrease of wood wettability demonstratedby the higher contact angles measured after treatmenthas in a first time no or limited effects on the high moisturecontent recorded after biological test confirming resultspreviously reported indicating that hydrophobic characterconferred to wood after thermal treatment was not at theorigin of the improvement of durability (Kamdem et al.,2002; Gosselink et al., 2004; Hakkou et al., 2005, 2006).
Durability of G. robusta heat treated wood was theninvestigated with different brown rot and white rot fungi.Tests were carried out with two European species (C. versi-
color and P. placenta) and two tropical species (P. sanguin-
eus and Antrodia sp.). Results are presented in Fig. 6. In allcases heat treated G. robusta was resistant to fungal attack.However, these results must be moderated according to thetested fungus. Indeed, while untreated G. robusta samplesare strongly degraded by C. versicolor and Antrodia sp.,heat treated samples show no degradation showing the effi-ciency of the treatment. Results are not significant in thecase P. placenta and P. sanguineus, for which untreatedsamples like heat treated ones were not or only slightly
Table 1Effect of time of treatment at 250 �C on durability of Grevillea robusta
wood after three months exposure to Coriolus versicolor
Sample h(�)
ML(%)
MC(%)
WL(%)a
Grevillea robusta heat treated0.5 h
113 12 111 3.6 ± 0.5
Grevillea robusta heat treated 1 h 116 16 104 2.2 ± 0.6Grevillea robusta heat treated 5 h 113 20.5 88 0Grevillea robusta heat treated 7 h 118 22 78 0Grevillea robusta heat treated
15 h111 26 49 0
Grevillea robusta control 31 – 97 25.8 ± 11.2Fagus sylvatica control – – 53 33.5 ± 3.4
a Average value on four replicates.
Coriolus versicolor (Cv), Poria placenta (Pp), Pycnoporus sanguineus (Ps), Antrodia sp.(Asp)
-5
0
5
10
15
20
25
30
35
40
Cv Pp Ps Asp
Fungal Species
Wt L
oss
(%)
Heat Treated Sample Control
Fig. 6. Weight losses of Grevillea robusta heat treated or not for 7 h at250 �C after three months exposure to different fungi.
0
2
4
6
8
10
12
14
Wei
ght l
oss
(%)
Untreated Grevillea robusta
Heat treated Grevillea robusta
Untreated Pinus sylvestris
Fig. 7. Weight losses of Grevillea robusta heat treated for 7 h at 250 �Cand controls tested against termites in laboratory conditions for 28 days.
3484 F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486
attacked. Such a behaviour could probably be explained bywater logging of the wood blocks hence not allowing thedevelopment of the fungi.
Results concerning resistance to termites after 28 days inlaboratory conditions are presented in Fig. 7. Heat treatedG. robusta showed insignificant termite attack, whileuntreated samples were severely attacked (compare weightlosses of less than 1 for control to weight losses comprisedbetween 11% and 14% for heat treated samples). UntreatedG. robusta is more susceptible to termites than untreatedP. sylvestris. Survival rate of termites was high in the con-trols. Mortality rate was high in the treated samples as theresult of feeding on treated wood or inability to feed on ithence starve to death. After three weeks, all the termites inthe treated sample containers were noted to have died.These results were confirmed under field test conditions(Fig. 8). There was significant difference between attackson the heat treated G. robusta samples and the controlswhen exposed to termites in the natural environments.Attack on all samples increased with exposure time.
4. Conclusion
This study shows that heat treatment of G. robustaincreases its durability against basidiomycetes and termites.The decay resistance depends of the treatment conditionand is total for treatments performed at 250 �C for 7 h.Wettability changes observed after the heat treatment can-not be used to explain wood enhanced durability. Similarlyto studies reported on non durable European wood species,chemical modifications seems to be the most plausiblehypothesis to explain wood durability improvement. Masslosses, 13C MAS NMR and FTIR analysis indicate impor-tant degradation of hemicelluloses. Moreover, NMR ana-lysis indicate the presence of degradation products withinthe wood polymeric components due to a beginning of deg-radation of cellulose. Microscopic analysis indicate thatwood anatomy was slightly affected and that some shrink-age occurred after treatment. All these results show thatheat treatment could be a valuable alternative to toxicchemicals used actually to treat G. robusta wood in Kenya.This is confirmed by improved durability against fungi and
termites in the laboratory and in the field. However, ourresults needs further investigations to validate laboratoryresults on larger scale and to study the effect of heat treat-ment on mechanical properties.
Acknowledgement
This work was supported by grant from the Frenchgovernment through the Embassy in Nairobi.
References
Alen, R., Kotilainen, R., Zaman, A., 2002. Thermochemical behaviour ofNorway spruce (Picea abies) at 180–225 �C. Wood Sci. Technol. 36,163–171.
Cannon, J.R., Chow, P.W., Fuller, M.W., Hamilton, B.H., Metcalf, B.W.,Power, A.J., 1973. Phenolic constituents of Grevillea robusta (Prote-aceae). Structure of robustol, a novel macrocyclic phenol. Aust. J.Chem. 26 (10), 2257–2275.
Gosselink, R.J.A., Krosse, A.M.A., van der Putten, J.C., van der Kolk,J.C., de Klerk-Engels, B., van Dam, J.E.G., 2004. Wood preservationby low-temperature carbonisation. Ind. Crops Prod. 19, 3–12.
Hakkou, M., Petrissans, M., El Bakali, I., Gerardin, P., Zoulalian, A.,2005. Wettability changes and mass loss during heat treatment ofwood. Holzforschung 59 (1), 35–37.
Hakkou, M., Petrissans, M., Gerardin, P., Zoulalian, A., 2006. Investi-gation of the reasons for fungal durability of heat treated beech wood.Polym. Degrad. Stab. 9 (2), 393–399.
Hill, C., 2005. The Use of timber in the twenty-first century. In: WoodModification, Chemical Thermal and other Processes, Wiley Series inRenewable Resources, pp. 1–18, ISBN-10 0-470-02172-9.
Kamdem, D.P., Pizzi, A., Jermannaud, A., 2002. Durability of heat-treated wood. Holz als Roh- und Werkstoff 60, 1–6.
KFMP, Kenya Forest Master Plan, 1994. Development Programmes,Ministry of Environment and Natural Resources, Government ofKenya.
Mouras, S., Girard, P., Rousset, P., Permadi, P., Dirol, D., Labat, G.,2002. Proprietes physiques de bois peu durables soumis a un traitementde pyrolyse menagee. Ann. Forest Sci. 59, 317–326.
Nuopponen, M., Vuorinen, T., Jasma, S., Viitaniemi, P., 2004. Thermalmodifications in softwood studied by FT-IR and UV resonanceRaman spectroscopies. J. Wood Chem. Technol. 24 (1), 13–26.
Patzelt, M., Stingl, R., Teischinger, A., 2002. Termische Modifikation vonHolz und deren Einflub auf ausgewahlte Holzeigenschaften, In:Modifiziertes Holz Eigenschaften und Markte, LIGNOVISIONENBand 3, pp. 101–149, ISSN 1681–2808.
Ridley, D.D., Ritchie, E., Taylor, W.C., 1968. Chemical studies of theProtaceae II. Constituents of Grevillea robusta; the synthesis of n-tridecylresorcinol (grevillol) and related substances. Aust. J. Chem. 21(12), 2979–2988.
Ritchie, E., Taylor, W.C., Vautin, S., Protaceae, I., 1965. Grevillearobusta and Orites excelsa. Aust. J. Chem. 18 (12), 2015–2020.
Santos, J.A., 2000. Mechanical behaviour of eucalyptus wood modified byheat. Wood Sci. Technol. 34, 39–43.
Sivonen, H., Maunu, S.L., Sundholm, F., Jamsa, S., Viitaniemi, P., 2002.Magnetic resonance studies of thermally modified wood. Holzfors-chung 56, 648–654.
Standard method laboratory for evaluation to determine resistance tosubterranean termites, 1997. American Wood Preserver’s AssociationStandard, E1-97, p. 4.
Standard method of evaluating wood preservatives by field tests withstakes, 1993. American Wood Preserver’s Association Standard, E1-97, p. 9.
Tjeerdsma, B.F., Boonstra, M., Pizzi, A., Tekely, P., Militz, H., 1998.Characterisation of the thermally modified wood: molecular reasons
0
5
10
15
20
25
30
35
40
Time (months)
Wei
ght L
oss
(%) Heat treated Grevillea
robusta
Untreated Grevillearobusta
Untreated Pinussylvestris
0 1 2 3 4
Fig. 8. Weight losses of Grevillea robusta heat treated for 7 h at 250 �Cand controls tested against termites in field conditions.
F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486 3485
for wood performance improvement. Holz Roh-Werkstoff 56, 149–153.
Unsal, O., Ayrilmis, N., 2005. Variations in compression strength andsurface roughness of heat treated Turkish river gum (Eucalyptus
camaldulensis) wood. J. Wood Sci. 51, 405–409.Weiland, J.J., Guyonnet, R., 2003. Study of chemical modifications and
fungi degradation of thermally modified wood using DRIFT spec-troscopy. Holz Roh-Werkstoff 61, 216–220.
Wikberg, H., Maunu, S.L., 2004. Characterisation of thermally modifiedhard- and softwoods by 13C CPMAS NMR. Carbohyd. Polym. 58,456–461.
Yildiz, S., Gezer, E.D., Yilditz, U.C., 2006. Mechanical and chemicalbehavior of spruce wood modified by heat. Build. Environ. 41, 1762–1766.
Zawadzki, J., Wisniewski, M., 2002. 13C NMR study of cellulose thermaltreatment. J. Anal. Appl. Pyrolysis 62, 11–121.
3486 F. Mburu et al. / Bioresource Technology 98 (2007) 3478–3486
Appendix 3
Study and valorization of different Kenyan wood species
Abstract : The first part concerns the study of the reasons of natural durability of Prunus africana. Extractions were performed on
heartwood and sapwood using different solvents and the extracts tested. The aim was to test the effect of extractives against
fungi and termites in relation to the natural durability of Prunus africana. These were tested for mycelium growth inhibition
experiment against Coriolus versicolor, Poria placenta, and Aureobasidium pullulans at concentrations of 50 ppm, 100 ppm,
500 ppm and 1000 ppm. Its extracts showed high inhibition rate against fungi at low concentrations. Inhibition increased with
extract concentration and decreased with time. In some instances, fungal growth started after the control Petri dish was fully
colonized showing rather a fungistatic effect than a fungicidal one. Extracted and un-extracted wood samples were also tested
against termites. When tested against Coriolus versicolor in the laboratory Prunus africana showed high natural durability.
Extracted wood showed low resistance against termites while un-extracted was totally resistant. Dichloromethane extract
showed the highest inhibition rate compared to water, acetone and toluene/ethanol extracts against fungi and termites.
Microscopic analysis indicate the presence of important quantities of extractives deposited in the vessels which are removed
after Soxhlet extraction. Spectroscopic and chromatographic analysis were investigated on dichloromethane extract to identify
the compounds responsible for biological activities.
The second part of the study concerns the durability improvement of Grevillea robusta, a tropical wood species of low natural
durability. Grevillea robusta was treated by cheap sap displacement method using Copper Chrome Arsenate (CCA). Results
showed adequate penetration and retention for wood use in different hazard areas as recommended by Food and Agricultural
Organization (FAO). Analysis of wood components showed low lignin quantity and high holocelluloses content.
Grevillea robusta was heat treated under inert conditions to improve its durability. Resistance of heat-treated samples was
evaluated by malt agar block test against six wood rotting fungi as well as against termites in laboratory and field conditions.
High durability against fungi and termites was realized after treatment. There was a positive correlation between decay
resistance and mass loss due to thermal treatment. FTIR and 13C MAS NMR analysis showed chemical modification of wood
components after heat treatment.
Key words: Biodeterioration, CCA, decay, durability, fungi, Grevillea robusta, heat treatment, Prunus africana, termites
Etude et valorisation de différents bois du Kenya
Résumé :
La première partie de ce travail est consacrée à l’étude des raisons de la durabilité naturelle du Prunus africana. Des
extractions ont été réalisées à partir du duramen et de l’aubier avec différents solvants et les propriétés antifongiques et anti-
termites des extraits étudiées. Des essais d’inhibition de croissance du mycélium ont été réalisés sur Coriolus versicolor, Poria
placenta et Aureobasidium pullulans à plusieurs concentrations, 50, 100, 500 et 1000 ppm. Les résultats mettent en évidence
une importante inhibition de croissance des différentes souches fongiques même à faible concentration. L’inhibition augmente
avec la concentration de la solution testée et diminue avec le temps. Dans certains cas, la croissance du champignon débute
après que la boîte de Pétri témoin ait totalement été colonisée, indiquant plutôt un effet fongistatique que fongicide. Dans le
même temps, des essais réalisés sur éprouvettes de bois en présence de Coriolus versicolor dans des conditions de laboratoire
indiquent une forte durabilité naturelle du duramen. La résistance d’échantillons de bois extraits ou non a également été
évaluée en présence de termites. Les échantillons extraits présentent une faible durabilité contre ces dernières alors que les bois
non extraits sont totalement résistants. Les composés présents dans les extraits obtenus à l’aide du dichlorométhane présentent
une plus grande activité que ceux obtenus à l’eau, avec de l’acétone ou un mélange toluène/éthanol aussi bien au niveau des
termites que des champignons. Des analyses microscopiques indiquent la présence d’importantes quantités d’extractibles
présents dans les vaisseaux qui disparaissent après extraction au Soxhlet. Des analyses spectroscopiques et
chromatographiques sont réalisées sur les extraits dichlorométhane pour identifier les composés responsables de l’activité
biologique.
La seconde partie du travail concerne l’amélioration de la durabilité du Grevillea robusta, une essence tropicale de faible
durabilité naturelle. Dans un premier temps, l’imprégnabilité du Grevillea robusta par une solution de CCA, a été évaluée en
utilisant un procédé peu coûteux de déplacement de sève. Les résultats obtenus indiquent une bonne pénétration et rétention du
produit dans le bois permettant l’utilisation de ce dernier dans différentes classes de risques en fonction de la concentration de
la solution d’imprégnation et des recommandations de la FAO (Food and Agricultural Agency). Dans un second temps,
l’amélioration de la durabilité du Grevillea robusta a été envisagée en utilisant un traitement thermique sous atmosphère inerte.
La résistance des blocs traités thermiquement a été évaluée sur milieu gélosé sur six champignons de pourriture ainsi que sur
les termites en conditions de laboratoire et de champ. Dans tous les cas, une forte augmentation de la durabilité est observée
après traitement. Il existe une forte corrélation entre la perte de masse due au traitement thermique et la résistance à la
dégradation fongique. Des analyses IR et RMN indiquent d’importantes modifications des constituants polymériques du bois
après traitement thermique. Les différents résultats obtenus montrent que différemment la durabilité du Grevillea robusta peut
être facilement améliorée, permettant son utilisation comme alternative à celle d’essences durables surexploitées à ce jour au
Kenya.
Mots clés : Biodégradation, CCA, durabilité, Grevillea robusta, traitement thermique, pourriture, Prunus africana, termites