Available via license: CC BY 4.0
Content may be subject to copyright.
Nakaietal. J Wood Sci (2020) 66:67
https://doi.org/10.1186/s10086-020-01915-x
ORIGINAL ARTICLE
Flow deformation characteristics ofAfrican
blackwood, Dalbergia melanoxylon
Kazushi Nakai1,2* , Soichi Tanaka2, Kozo Kanayama2 and Tsuyoshi Yoshimura2
Abstract
African blackwood (ABW: Dalbergia melanoxylon) is a valuable tree in Tanzanian local community forests, and heart-
wood has been mainly utilized as an irreplaceable material in musical instruments, e.g., clarinet, oboe and piccolo.
Since its use is generally for the production of musical instruments only, most of the harvested volume is wasted due
to defects that would affect the quality of final products. Wood flow forming can transform bulk woods into materials
in temperature/pressure-controlled mold via plastic flow deformation. The main object of this study was to evaluate
the deformation characteristics of ABW heartwood in developing the potential of wasted ABW parts in terms of the
effective material use. The deformation characteristics of heartwood were examined by free compression tests. Speci-
mens were compressed along the radial direction at 120 °C, and air-dried heartwood was dramatically deformed in
the tangential direction. The plastic flow deformation of ABW was amplified by the presence of both extractives and
moisture. In particular, the ethanol/benzene (1:2, v/v) soluble extractives in heartwood may have contributed to flow
deformation. The results of the dynamic mechanical analysis showed that the air-dried heartwood exhibited softening
in a temperature range over 50 °C. The ethanol/benzene-soluble extractives contributed to the softening behavior.
The clarified deformation characteristics of ABW can contribute to more efficient material use of local forests.
Keywords: Dalbergia melanoxylon, African blackwood, Wood flow forming, Plastic flow deformation, Dynamic
viscoelasticity
© The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material
in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material
is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the
permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco
mmons .org/licen ses/by/4.0/.
Introduction
African blackwood (ABW), Dalbergia melanoxylon, is
the national tree of Tanzania, called “Mpingo” in Swahili.
e tree naturally occurs in the dryland of sub-Saharan
Africa, including in eastern African countries such as
Tanzania and Mozambique. ese two countries actually
stock intensive natural resources of this tree, and they
have recently provided most ABW timber. In Tanzania,
ABW trees tend to be observed in a semi-deciduous for-
mation including deciduous and evergreen trees, i.e., the
miombo woodlands, which are characterized by an abun-
dance of three genera: Brachystegia, Julbernardia and Iso-
berlinia [1, 2]. e trees naturally grow in clusters, with a
population density that has been estimated as 9–90 trees/
ha [3–6].
It is mostly the heartwood of ABW that is utilized in
the production of musical instruments, especially wood-
winds, e.g., clarinets, oboes, piccolos, and bagpipes, due
to its specific characteristics. It is normally purplish-
black in color, and extremely heavy, having an air-dried
density of 1.1–1.3g/cm3 [7, 8], while loss tangent (tanδ)
in vibration properties is lower than other general hard-
wood species [9]. ese characteristics are greatly differ-
ent from those of the milky-white sapwood, which has an
air-dried density of 0.75g/cm3 [7].
We recently reported that the growth characteris-
tics of ABW are affected by the surrounding environ-
ment, i.e., topography, climate and human activities.
Nevertheless, the tree can survive under various envi-
ronmental conditions with intensive population [6].
Several defects that frequently occur in natural trees,
such as lateral twists, deep fluting, and knots including
Open Access
Journal of Wood Scienc
e
*Correspondence: kazushi.nakai@music.yamaha.com; kazushi_nakai@rish.
kyoto-u.ac.jp
2 Research Institute for Sustainable Humanosphere, Kyoto University,
Gokasho, Uji, Kyoto 611-0011, Japan
Full list of author information is available at the end of the article
Page 2 of 11
Nakaietal. J Wood Sci (2020) 66:67
cracks, can affect the operation of sawmills [10]. ey
also influence the properties of musical instruments.
us, sawmills can produce only a small amount of the
necessary quality timber, with an actual timber yield of
only 9% [11]. Currently, ABW is traded as one of the
most high-priced timbers in the world; meanwhile, this
inefficient utilization has threatened the species’ future
existence [11, 12]. In Tanzania, wasted ABW is fre-
quently used as an energy resource, e.g., charcoal and
fuelwood. As such, it is sold at prices very much lower
than the timber [13]. us, additional uses for ABW
waste could potentially contribute to the development
of local communities.
Wood flow forming is a novel technique for molding
three-dimensional products from bulk wood [14]. It has
been suggested that wood flow deformation is plastic
flow deformation due to the slipping of wood cells in or
around the intercellular layers under specific tempera-
tures [15]. Plastic flow deformation occurs under high
compressive load after two types of compressive defor-
mation stages: elastic deformation and densification
deformation [15, 16]. e bulk wood of air-dried Japa-
nese cedar (Cryptomeria japonica) was deformed and
flowed in a mold by the addition of both high pressure
and high temperature, i.e., 150kN loading at 130°C [17].
Furthermore, flow deformation could be promoted by
increasing moisture content and/or adding thermosetting
polymers [15, 17–19]. In particular, it has been suggested
that an increase in the polymer content can dramatically
improve flow deformation [20, 21]. Timber impregnated
with polymer can flow based on the thermal softening
behavior of the polymer.
e quantity of extractives in ABW heartwood has
been estimated to be over 15 wt% in ethanol/benzene
(1:2 v/v) solvent extraction, which is much higher than
in other Dalbergia species, such as Dalbergia cultrate
and Dalbergia latifolia [22]. e high concentrated
extractives potentially work to promote flow deforma-
tion by heating beyond the thermal softening point of
them. Although identification and isolation of extrac-
tives obtained from ABW heartwood have been partly
reported [22–25], there is little information about the
effect of extractives on the thermal behavior of ABW.
e main objective of this study was to reveal the
deformation characteristics of ABW, and to discuss the
relationship between flow deformation of heartwood and
thermal behavior of extractives. To examine the deforma-
tion characteristics of ABW, we conducted free compres-
sion tests. In addition, the thermal behavior of extractives
was evaluated based on the temperature dependence of
the dynamic viscoelasticity of ABW in the modulus of
transverse elasticity. We mainly focused on the heart-
wood which occupies the large part of ABW trees, so that
the results might contribute to improvement of the gen-
eral material utilization of ABW.
Materials andmethods
Wood specimens
Specimens of ABW were obtained from logs over 24cm
in diameter at breast height, harvested in 2018 at the For-
est Stewardship Council (FSC)-certified forest located
in the Kilwa District, Lindi Region, Tanzania. Two types
of specimens were prepared for this study. Disk-shaped
specimens of ABW heartwood obtained from tangential
sections of the wood with 15-mm diameter (longitudi-
nal, L × tangential, T) and 2-mm thick in the radial direc-
tion were prepared for the free compression test with 20
replicates. Rectangular specimens of ABW (heartwood
and sapwood), measuring 30mm (L; longitudinal direc-
tion) × 1 mm (R; radial direction) × 5 mm (T; tangential
direction), were prepared for the dynamic mechanical
analysis with 15 replicates per specimen. All specimens
were cut from air-dried timber conditioned for over
3months at room temperature, and specimens were kept
in a controlled chamber (KCL-2000, Tokyo Rikakikai Co.
Ltd., Tokyo, Japan) conditioned at 22 ± 2°C and 60% rela-
tive humidity (RH) for over 30days.
Pretreatment prior tothetests
Figure1a, b shows the experimental procedures for the
disk-shaped and rectangular specimens, respectively.
For both, four types of treatment [air-drying (AD), water
extraction (WT), ethanol/benzene extraction (EB), and
oven-drying (OD)], were prepared with 5 replicates
according to the experimental procedures (Fig.1a, b). All
specimens were oven-dried at 105°C for over 24h, and
their oven-dried weights (W0) were measured with an
electronic scale (GH-252, A&D Company Ltd., Tokyo,
Japan).
e extraction processes were applied for the WT
and EB specimens (Fig. 1). Water extraction was per-
formed as follows. Oven-dried specimens were soaked in
150mL of distilled water using a sealed Erlenmeyer flask.
e soaked specimens were stirred for 10 min in the
water bath at 40°C with ultrasonic treatment (Branson
5510JDTH, Yamato Scientific Co., Ltd., Tokyo, Japan),
and then kept in the controlled chamber at 45 ± 5 °C for
48h. For the extraction, specimens from different sam-
pling parts (heartwood, sapwood) were placed in differ-
ent flasks to prevent the migration of extractives between
parts. For the EB specimens, extraction was performed
in the same way using an (1:2, v/v) ethanol/benzene solu-
tion instead of water.
After extraction, both the WT and EB specimens were
stored at room temperature for over 1 week, and then
oven-dried at 105°C for over 24h to measure the extracted
Page 3 of 11
Nakaietal. J Wood Sci (2020) 66:67
weight (We) with the electric scale (Fig.1a, b). e extrac-
tion rate was calculated by Eq.1 using W0 and We:
(1)
Extraction rate
=
W
0−
W
e
W
0
×100(%)
.
e AD and extracted WT and EB specimens were
conditioned at 22 ± 2°C and 60% RH for over 3weeks.
e conditioned weight (W1) was then measured with
the electric scale. In addition, the dimensions of each
specimen were measured after the conditioning process,
as described later. e moisture content (MC) of each
Fig. 1 Experimental procedures for: a free compression test, 5 replicates of disk-shaped specimen for AD, WT, EB and OD; and b for dynamic
mechanical analysis (DMA), 5 replicates of rectangular specimen for AD, WT and EB
Page 4 of 11
Nakaietal. J Wood Sci (2020) 66:67
specimen was calculated before the tests using the fol-
lowing Eqs.2a and 2b:
e MC of AD specimens was calculated using Eq.2a,
while that of extracted specimens (WT and EB) was cal-
culated using Eq.2b.
e dimensions of AD, WT, EB, and OD specimen
were measured just before the tests (Fig.1a, b). For the
specimens provided to free compression test (Fig.1a), the
dimension of radial direction (R-direction, h0) was meas-
ured at the center point of specimens with a micrometer
(OMV-25MX, Mitutoyo Corp., Kawasaki, Japan); the
dimensions of longitudinal (L-direction) and tangential
directions (T-direction) were measured at the center-
line of each direction with a digital caliper (CD-15CP,
Mitutoyo Corp., Kawasaki, Japan). e cross-sectional
area was calculated using the image processing software
ImageJ [26, 27]. For dynamic mechanical analysis (DMA)
specimens (Fig.1b), R-direction and T-direction dimen-
sions were measured at the centerline of each with the
above-noted digital caliper.
Free compression test
e free compression test was performed with a universal
testing machine (Instron 5582, Instron Co., MA, USA) as
illustrated in Fig.2. Specimens were placed on the lower
punch controlled at 120°C, and held in place with the
upper punch without loading for the pre-heating time of
60s. (Fig.2). ey were then compressed at a constant
speed (0.02mm/s), while both compressive stress (P) and
gap displacement caused by deformation of specimen (hs)
were measured. Compression was also performed with-
out specimens, the P and the gap displacement caused by
(2a)
MC
=
W
1
−W
0
W
0
×100(%)
,
(2b)
MC
=
W
1
−W
e
W
e
×100(%)
.
deformation of punches (hb) were measured. e actual
displacement (h) was calculated using Eq.3:
e stress–strain curve was described using nominal
strain (ε) and nominal stress (σ) calculated using Eqs.4
and 5:
where h0 is the initial specimen thickness (in the R-direc-
tion), π is the circular constant, and d is the diameter of
the punch (d = 15mm). Specimens were compressed to
a maximum compressive load of 20 kN, equivalent to
113MPa in compressive stress. In this study, water vapor
pressure, caused by heating air-dried specimens, was
neglected due to the small specimen size.
After the test, all specimens except for the OD were
placed in a controlled chamber for 1week at 22 ± 2 °C
and 60% RH for conditioning, and the parameters of
specimen weight, dimensions (L-direction, R-direction,
and T-direction) and cross-sectional area, were meas-
ured (Fig. 1a). e parameters of OD were measured
immediately after the test. Dimensional changes (Dc)
caused by the test (L-direction, T-direction and cross-
sectional area) were calculated by Eq.6:
where Db and Da are the dimensional values of specimens
before and after the test, respectively.
e physical parameters, Young’s modulus and maxi-
mum strain, were determined from the stress–strain
curve collected through the test results. Young’s modulus
(E) was calculated from the angle of elastic deformation
area in the curve. e stress at the flow-starting point
(σf) was defined as the inflexion point of the stress–strain
curve (Fig.3), where the first peak of the derivative stress
with respect to the strain (dσ/dε). e strain at the inflec-
tion point was defined as the flow-starting strain (εf).
e maximum strain (εm) was defined as the compres-
sive strain value at the maximum compressive stress,
σm = 113MPa in the test.
Dynamic mechanical analysis
e DMA was performed using a rheometer (ARES-
G2, TA Instruments, New Castle, USA). e complex
dynamic modulus (G*) of viscoelastic materials generally
represents the relation between the storage modulus (G′)
and loss modulus (Gʺ), which are calculated from the
(3)
h
=
hs
+
hb.
(4)
ε
=
1
−
(h/h0),
(5)
σ
=P
/(π
d
2/
4
),
(6)
D
c=
D
a−
D
b
D
b
×100(%)
,
Fig. 2 Schematic diagram of free compression test. d: specimen
diameter, h0: initial thickness, h: deformed thickness
Page 5 of 11
Nakaietal. J Wood Sci (2020) 66:67
dynamic performance with oscillation stress and strain
by Eqs.7 and 8:
where i is the imaginary number, ω is the angular fre-
quency, δ is the phase angle, and tanδ is the loss factor.
In this study, G′, Gʺ and tanδ were calculated from the
amplitude and phase difference (δ) of the oscillation
curve for torque using the analysis software (TRIOS, TA
Instruments, New Castle, USA).
e temperature-ramp test was conducted under
a controlled environment by N2 purge, from – 50 to
250°C at a constant temperature ramp rate (5°C/min).
Both edges of specimens were cramped at 20mm in the
L-direction, and loaded with dynamic torsion, 0.5% oscil-
lation shearing strain at a constant frequency of 1.0Hz
(Fig.4).
Statistical analysis
e Tukey–Kramer test at 1% critical difference (p < 0.01)
was used to analyze statistical differences between values
(BellCurve for Excel, Social Survey Research Information
Co. Ltd., Tokyo, Japan).
Results anddiscussion
Deformation characteristics
In the free compression test, both AD and WT speci-
mens were flowed in the T-direction at 120°C, while EB
and OD specimens were not flowed (Figs. 5, 6). ese
(7)
G
∗=G
′
(ω)+iG
′′
(ω)=
G∗
(cos δ+isin δ)
,
(8)
tan δ
=G
′′
(ω)/
G
′
(ω),
findings suggest it was possible that flow deformation
was promoted by the extractives and moisture.
In this study, AD showed the largest flow deforma-
tion compared to the other specimens. Table1 lists the
extraction rate, initial MC, and dimensional change for
both the L- and T-directions together with specimens’
cross-sections. In the cross-sectional area, AD was again
the highest (average ca. 117%) of all specimens with a sig-
nificant difference at 1% level. For AD, the dimensional
change in the T-direction was also the highest among
specimens, whereas those in L-direction had no signifi-
cant difference. ese dimensional values in the vertical
direction of compression loading show the displacement
caused by flow deformation. e results suggest that the
changes in the T-direction corresponded strongly to flow
deformation based on the wood anisotropy: the lower
strength on T-direction than L-direction. Yamashita etal.
[17] found that flow direction was mainly perpendicular
to the fiber orientation, which was in keeping with our
findings.
Figure7a, b shows the σf and εf values for each speci-
men, the stress and strain values specialized at the flow-
starting point. e σf value indicates the stress value
necessary to generate flow deformation. AD showed the
lowest value (average ca. 33.0 MPa) of all specimens,
while WT, EB and OD values were significantly higher
(Fig. 7a). e εf value indicates the strain required to
initiate flow deformation. ere was a significant differ-
ence between the non-extracted specimens (AD and OD)
and the extracted specimens (WT and EB) at 1% level
(Fig.7b).
Figure8 shows the maximum strain (εm) values. e εm
value reflects the total displacement by the loaded com-
pressive stress, i.e., elastic deformation, densification
Fig. 3 Method for calculating the deformation parameters from the
stress–strain curve
Fig. 4 Experimental set-up for the dynamic mechanical analysis
using a rheometer
Page 6 of 11
Nakaietal. J Wood Sci (2020) 66:67
deformation, and flow deformation. Here, there were sig-
nificant differences among all specimens at 1% level: ca.
73% (AD), ca. 58% (WT), ca. 28% (EB) and ca. 10% (OD)
on average. e difference in the strain between εm and
εf (Δε = εm– εf), which indicated the displacement due
to flow deformation, showed a significantly high value in
AD (ca. 58.4%) at 1% level (Figs.7b, 8). Although there
was no statistical difference among the values of other
specimens, the variation depended on extractives and
moisture: ca. 21.2% (WT), ca. 0.80% (EB) and ca. 1.30%
(OD) (Figs.7b, 8). erefore, it was suggested that extrac-
tives and moisture amplified flow deformation.
e ethanol/benzene extractives appeared to influence
the deformation characteristics. Young’s modulus (E) of
specimens is shown in Fig. 9. e E values of WT and
EB were statistically similar to that of AD at the 1% level,
despite their average values being more than 2 times
higher: 300.9MPa (AD), 641.5MPa (WT) and 999.6MPa
(EB). e MC of EB was also similar to those of AD and
WT (Table1). is suggested that the ethanol/benzene-
soluble extractives likely increased the elastic deforma-
tion of ABW. e σf value in EB was the highest among
all specimens, although statistically equal to OD (Fig.7a).
e εf value in EB was higher than AD (Fig.7b), although
the εm value in EB was significantly lower than WT
and AD with only 1% of dimensional change (Table 1,
Fig.8). e value of Δε in EB was also significantly lower
than AD (Figs.7b, 8). ese findings suggested that the
Fig. 5 Specimen in the free compression test: a before and b after compression up to 113 MPa compressive stress
Fig. 6 Representative change in shapes of specimens before and
after the free compression test
Table 1 Extraction rate, moisture content (MC) anddimensional changes due tofree compression (mean ± SD)
* Means with the same letter (a, b, c) are not signicantly dierent (Tukey–Kramer test, p < 0.01) (n = 5)
Specimens Extraction rate %* Moisture content %* Dimensional change %
Cross-sectional area* L-direction* T-direction*
AD 8.24 ± 0.61a117.39 ± 7.90c− 0.52 ± 1.21a97.45 ± 10.00c
WT 1.89 ± 0.44a8.88 ± 1.22a83.02 ± 3.38b3.54 ± 4.54a66.06 ± 24.24b
EB 16.12 ± 1.23b8.69 ± 0.11a1.01 ± 1.20a− 0.51 ± 1.46a1.13 ± 0.63a
OD 1.15 ± 0.86a0.61 ± 0.55a2.29 ± 0.85a
Page 7 of 11
Nakaietal. J Wood Sci (2020) 66:67
ethanol/benzene-soluble extractives helped promote the
flow deformation of ABW.
e significant improvement of wood plastic defor-
mation by increasing resin content has been reported
previously [19–21, 28]. Wood extractives are gener-
ally distributed in cell walls and intercellular layers, as
well as in lumen. erefore, the presence of extractives
potentially influences plastic deformation, despite their
small amounts. In this study, EB showed a significantly
higher extraction rate than WT: 16.12% (EB) and 1.89%
(WT) on average (Table1). e ethanol/benzene-soluble
extractives comprising over 16wt% of ABW heartwood
(Table1) [22], apparently have a large impact on ABW
deformation characteristics.
e water-soluble extractives might also influence
deformation characteristics. Dimensional change in the
cross section of WT was significantly lower than that
of AD: average ca. 83% (Table1). e σf value in WT
was higher than AD at 1% level: average ca. 80.9 MPa
(Fig.7a). e value of Δε in WT was also significantly
lower, as previously noted. Furthermore, an increase of E
in WT was observed though the value was not statisti-
cally different from AD (Fig.9). erefore, the presence
of water-soluble extractives also contributed to promot-
ing the flow deformation of ABW. ese results might
also indicate a positive relationship between water-solu-
ble extractives and the deformation loading required to
generate flow deformation.
e moisture in wood also affected the deformation
characteristics. As shown in Table1, flow deformation
was not observed in OD, even though extractives were
present. e σf value in OD was approximately three
times higher than in AD (Fig.7a), and Δε values were also
0
20
40
60
80
100
120
AD WT EB OD
σf(MPa)
Samples
(a)
a
bcc
0.0
0.2
0.4
0.6
0.8
1.0
AD WT EB OD
εf
Samples
(b)
a
b
b
a
Fig. 7 Stress and strain at flow-starting point. a Compressive stress at the flow-starting point (σf). b Compressive strain at the flow-starting point (ɛf).
Means with the same letter (a, b, c) are not significantly different (Tukey–Kramer test, p < 0.01) (n = 5). Error bars represent standard deviations
0.0
0.2
0.4
0.6
0.8
1.0
AD WT EB OD
εm
Samples
a
b
c
d
Fig. 8 Maximum strain at the maximum load (ɛm). Means with the
same letter (a, b, c, d) are not significantly different (Tukey–Kramer
test, p < 0.01) (n = 5). Error bars represent standard deviations
0
1000
2000
3000
4000
5000
6000
AD WT EB OD
E(MPa)
Samples
aa
a
b
Fig. 9 Young’s modulus (E) of specimens in the compressive fluidity
tests. Means with the same letter (a, b) are not significantly different
(Tukey–Kramer test, p < 0.01) (n = 5). Error bars represent standard
deviations
Page 8 of 11
Nakaietal. J Wood Sci (2020) 66:67
lower than AD (Figs.7b, 8). Furthermore, the E value of
OD was markedly higher than the others (Fig. 9). is
suggested that the deformation characteristics of ABW
heartwood were affected by moisture. e moisture
content influenced the elastic deformation of ABW, and
possibly amplified even plastic deformation. Previous
reports also noted that moisture contributed to wood-
softening behavior, and that wood flow deformation
could be improved in proportion to the increase in MC
[17, 21]. It is possible that the lower value of E contrib-
uted to densification deformation and flow deformation,
while the change in E depended on the moisture content
of specimens.
Temperature dependence ofdynamic viscoelasticity
Table2 presents the extraction rate and MC of specimens
used for the DMA. e EB-heartwood had a significantly
high extraction rate (average ca. 11.3%) compared to
other specimens, including the EB-sapwood (average ca.
2.4%). By contrast, the rates in the WT-heartwood and
sapwoods were similar. erefore, the ethanol/benzene-
soluble extractives were definitely concentrated in the
heartwood. Meanwhile, the MCs of heartwoods (8–9%)
were a bit lower than those of sapwoods (9–11%), and the
extractions showed no effect. e MC of EB-sapwood
was statistically equal to that of heartwood, which was
significantly lower than the other sapwood specimens.
Extractives have been suggested to affect the sorption
properties of wood [29–31], and the removal of extrac-
tives could result in an increase in swelling–shrinkage
behavior [32–34]. Water-soluble extractives have been
reported not to affect the wood sorption properties [34,
35]. Our results did not show any clear effects of the
extractives on the sorption properties of ABW.
Figure 10 shows the temperature dependence of the
dynamic viscoelastic parameters of AD, WT and EB
specimens. e curves of heartwood specimens were
obviously shifted toward flattened curves in a range over
50°C due to the extractions, whereas the curves of sap-
woods were not shifted. Amorphous polymers such as
lignin and hemicellulose generally influence the tem-
perature dependence of dynamic viscoelasticity in wood,
with variable performance related to MC [36–38]. In this
study, the modulus of transverse elasticity obtained by
the DMA depended on the extractives, and were relevant
to the flow deformation characteristics.
e Gʹ values of all heartwood specimens overlapped
from 120–130 to 250°C (Fig.10a). From −50 to 120–
130°C, AD, WT and EB-heartwood specimens showed
a similar pattern of curves, with a sharp decrease after
50 °C. e values were highest in AD specimens, fol-
lowed by WT, and the lowest in EB specimens. By con-
trast, all Gʹ curves overlapped from –50 to 250°C in the
sapwood specimens. Different trends of Gʹ curves were
found between heartwood and sapwood in the range of
50°C to 120–130°C. Multiple inflection points could be
clearly observed in the AD-heartwood specimens, but
the AD-sapwood specimens exhibited a single inflection
point in this range.
e increase in Gʺ and tanδ generally indicates an
increase in the viscosity of a material, which might be
related to deformation characteristics. Since we noted the
large flow deformation in the AD-heartwood, our results
suggested that the patterns of Gʺ and tanδ curves are cor-
related with flow deformation in regard to the extractives,
as discussed above (Figs.6, 7, 8a, Table1). e Gʺ curves
showed that the AD-heartwood had multiple shoulder
peaks, and higher peak values in the range of 50–150°C,
suggesting that the viscosity of AD-heartwood was also
increased, because the rapid increase of tanδ was simul-
taneously observed in this range (Fig.10a). Although the
similar patterns were observed in the curves of WT- and
EB-heartwood, the values were decreased by the extrac-
tion. e Gʺ value of EB-heartwood was the lowest of all
the heartwoods, and its shifted curve almost overlapped
that of sapwood specimens (Fig. 10a, b). e curves of
sapwood specimens did not shift through the extrac-
tions with lower values in the range. As a result, sapwood
might not show flow deformation like EB-heartwood
in the free compression test. In addition, the potential
of flow deformation under other temperatures was also
assumed based on the curves of Gʺ and tanδ. e flow
deformation of ABW heartwood was observed at 120°C
in this study (Figs.5, 6). e results in Fig.10a suggest
that the AD-heartwood potentially flowed under temper-
atures lower than 120°C due to the significant increase
in Gʺ and tanδ in the range over 50°C. Lower tempera-
ture should be useful not only for preserving the original
mechanical properties, but also for controlling viscosity
in the mold.
Extractives were suggested to affect the softening
temperature of the wood, which has particularly large
amounts of ethanol/benzene extractives, like those
reported in pao rosa (Swartzia fistuloides) [39]. e
sharp increases of Gʺ and tanδ in the AD-heartwood
Table 2 Extraction rate and MC of specimens
inthedynamic mechanical analysis (mean ± SD)
* Means with the same letter (a, b) are not signicantly dierent (Tukey–Kramer
test, p < 0.01) (n = 5)
Specimen Extraction rate %* Moisture content %*
Heartwood Sapwood Heartwood Sapwood
AD 8.01 ± 0.73a10.83 ± 0.76b
WT 2.17 ± 0.78a3.67 ± 1.34a8.83 ± 0.84a10.52 ± 0.59b
EB 11.32 ± 3.54b2.36 ± 1.24a8.72 ± 0.50a9.13 ± 0.64a
Page 9 of 11
Nakaietal. J Wood Sci (2020) 66:67
at 50–80 ℃ may indicate the softening behavior of
extractives. e lowest values of both Gʺ and tanδ were
observed in the EB-heartwood (Fig.10a). Although all
the tanδ curves of heartwoods were essentially over-
lapped in the range under 50°C, the sharp increase of
tanδ showed a significant increase of Gʺ in this range.
is suggested the ethanol/benzene-soluble extractives
were softened in the range over 50°C. e curves of
WT-heartwood also suggested that the water-soluble
extractives affected the dynamic viscoelasticity of AD-
heartwood (Fig.10a). e water extraction resulted in
ca. 30% reduction in the Gʺ value of WT-heartwood
at 120℃, and the static parameters of WT-heartwood
were statistically different from those of AD-heart-
wood (Figs.7, 8). is trend was observed only in the
heartwood specimens, even though the extraction
rate in WT-sapwood was same as the WT-heartwood
(Table2). Although further studies are needed to iden-
tify the effect of water-soluble extractives, we assumed
that some extractives were duplicated by the ethanol/
benzene extraction due to the similarity of solubility
parameters between ethanol and water [40].
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
G' (GPa)
AD
WT
EB
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
G'' (GPa)
0.00
0.05
0.10
0.15
0.20
-50 50 150250
tanδ
Temp. (
℃
)
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
G' (GPa)
AD
WT
EB
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
G'' (GPa)
0.00
0.05
0.10
0.15
0.20
-5050150 250
tanδ
Temp. (
℃
)
ab
Fig. 10 Temperature dependence of the dynamic viscoelasticity among extraction treatments in a heartwood and b sapwood specimens
Page 10 of 11
Nakaietal. J Wood Sci (2020) 66:67
Conclusions
e application of wood flow forming techniques in
ABW could contribute to developing the effective utili-
zation of wasted ABW timbers in the local forest sector.
e present study demonstrated the deformation char-
acteristics of air-dried ABW heartwood via the free com-
pression test. e air-dried heartwood of ABW flowed
at 120℃, our findings suggested that the extractives in
heartwood definitely resulted in flow deformation. e
flow deformation depended mainly on the ethanol/ben-
zene-soluble extractives, which were highly concentrated
in the heartwood. e ethanol/benzene-soluble extrac-
tives were suggested to be softened at temperatures over
50 ℃. e DMA results indicated that the increase in
Gʺ and tanδ were strongly related to flow deformation
in the free compression test; thus, the potential of flow
formation at other temperatures was also assumed in the
DMA. e flow deformation of ABW also depended on
MC, although the oven-dried heartwood did not flow
even with the presence of extractives. e MC affected
mechanical properties, and an increase in MC might
result in flow deformation. Consequently, our findings
suggest the possibility that wood flow forming might
contribute to further utilization of ABW timbers wasted
in local sawmill factories.
Abbreviations
ABW: African blackwood; tanδ: Loss tangent in vibration properties; FSC:
Forest Stewardship Council; AD: Air-drying; WT: Water extraction; EB: Ethanol/
benzene extraction; OD: Oven-drying; RH: Relative humidity ; W0: Oven-dried
weight; We: Extracted weight; W1: Conditioned weight; MC: Moisture content;
R: Radial; L: Longitudinal; T: Tangential; P: Compressive stress; hs: Gap displace-
ment caused by deformation of specimen; hb: Gap displacement caused by
deformation of punches; h: Actual displacement; ε: Nominal strain; σ: Nominal
stress; h0: Initial thickness of specimen; π: The circular constant; d: Diameter
of punch; DMA: Dynamic mechanical analysis; Dc: Dimensional change; Da:
Dimensional value before free compression test; Db: Dimensional value after
free compression test; E: Young’s modulus; σf: Stress at the flow-starting point;
dσ/dε: Derivative stress with respect to strain; εf: Flow-starting strain; εm: Maxi-
mum strain; σm: Maximum compressive stress; G*: Complex dynamic stress;
Gʹ: Storage modulus; Gʺ: Loss modulus; i: The imaginary number; ω: Angular
frequency; δ: Phase angle.
Acknowledgements
A part of this article was presented at 2018 SWST/JWRS International Conven-
tion, Nagoya, Japan, November 2018, and at the 70th Annual Meeting of
the Japan Wood Research Society, Tottori, Japan, March 2020. We thank Akio
Adachi, Research Institute for Sustainable Humanosphere, Kyoto University,
for his help in preparing test specimens. We also thank Makala Jasper and
Jonas Timothy, Mpingo Conservation & Development Initiative, Tanzania, for
their kind assistance in helping us understand local forests. We would like to
express our gratitude to Neil Bridgland and James Laizer, Sound & Fair Ltd., in
Tanzania, and Motoki Takata and Shoko Ishii, Yamaha Corporation for collect-
ing wood samples. Our appreciation also goes out to all the Tanzanian local
villagers for their kind support.
Authors’ contributions
KN designed the study, prepared wood samples, analyzed data and wrote the
manuscript. ST and KK assisted in data collection and contributed to interpre-
tation. ST, KK and TY critically reviewed the manuscript. All authors read and
approved the final manuscript.
Funding
This work was supported as a part of a joint study for fundamental research
on achieving sustainable forest utilization focusing on African blackwood
(Dalbergia melanoxylon) by the Yamaha Corporation, Research Institute for
Sustainable Humanosphere, Kyoto University and the Graduate School of
Agriculture, Kyoto University.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from
the corresponding author on reasonable request.
Competing interests
The authors declare that they have no conflict of interest.
Author details
1 Musical Instruments & Audio Products Production Unit, Yamaha Corpora-
tion, 10-1 Nakazawa-cho, Naka-ku, Hamamatsu 430-8650, Japan. 2 Research
Institute for Sustainable Humanosphere, Kyoto University, Gokasho, Uji, Kyoto
611-0011, Japan.
Received: 18 June 2020 Accepted: 14 September 2020
References
1. Campbell B, Frost P, Byron N (1996) Miombo woodlands and their use:
overview and key issues. In: Campbell B (ed) The miombo in transi-
tion: woodlands and welfare in Africa. Center for International Forestry
Research (CIFOR), Bogor, pp 1–10
2. White F (1983) The Zambezian regional centre of endemism. In: White
F (ed) The vegetation of Africa: a descriptive memoir to accompany the
UNESCO/AETFAT/UNSO vegetation map of Africa (Natural Resources
Research 20). UNESCO, Paris, pp 86–101
3. Gregory A, Ball SMJ, Eziefula UE (1999) Tanzanian Mpingo 98 full report.
Mpingo Conservation Project, Tanzania
4. Opulukwa MJ, Hamza KFS, Malende YHMB (2002) Inventory of Dalbegia
melanoxylon (Mpingo) in the southern part of Tanzania: the case of Nach-
ingwea. Afr Study Mono 23(1):1–10
5. Mariki AS, Wills AR (2014) Environmental factors affecting timber quality
of African Blackwood (Dalbergia melanoxylon). Mpingo Conservation &
Development Initiative, Kilwa Masoko
6. Nakai K, Ishizuka M, Ohta S, Timothy J, Jasper M, Lyatura NM, Shau V,
Yoshimura T (2019) Environmental factors and wood qualities of African
blackwood, Dalbergia melanoxylon, in Tanzanian Miombo natural forest. J
Wood Sci 65(1):39
7. Malimbwi RE, Luoga EJ (2000) Prevalence and standing volume of
Dalbergia melanoxylon in coastal and inland sites of southern Tanzania. J
Tropical For Sci 12(2):336–347
8. Sproßmann R, Zauer M, Wagenfur A (2017) Characterization of acoustic
and mechanical properties of common tropical woods used in classical
guitars. Res Phys 7:1737–1742
9. Brémaud I, El Kaïm Y, Guibal D, Minato K, Thibaut B, Gril J (2012) Charac-
terisation and categorisation of the diversity in viscoelastic vibrational
properties between 98 wood types. Ann For Sci 69(3):373–386
10. Lovett J (1987) Mpingo—the African blackwood. Swara 10:27–28
11. Jenkins M, Oldfield S, Aylett T (2002) International trade in African black-
wood. Fauna & Flora International, Cambridge
12. Hamisy WC, Hantula J (2002) Characterization of genetic variation in
African Blackwood, Dalbergia melanoxylon using random amplified
microsatellite (RAMS) method. Plant genetic resources and biotechnol-
ogy in Tanzania, part 1: biotechnology and social aspects. In: Proceed-
ings of the second national workshop on plant genetic resources and
biotechnology, Arusha, Tanzania, 6–10 May 2002
13. Miya M, Ball SMJ, Nelson FD (2012) Drivers of deforestation and forest
degradation in Kilwa District. Mpingo Conservation & Development
Initiative, Kilwa
14. Yamashita O, Yokochi H, Miki T, Kanayama K (2007) Producing cups from
wood by extrusion using flow phenomenon of bulk wood. Nihon Kikai
Ronbun 73(729):2074–2078 (in Japanese)
Page 11 of 11
Nakaietal. J Wood Sci (2020) 66:67
15. Miki T, Sugimoto H, Shigematsu I, Kanayama K (2014) Superplastic defor-
mation of solid wood by slipping cells at sub-micrometre intercellular
layers. Int J Nanotechnol 11(5–8):509–519
16. Sugino H, Tanaka S, Kasamatsu Y, Okubayashi S, Seki M, Miki T, Umemura
K, Kanayama K (2020) Influence of electron-beam irradiation on plastic
flow deformation of wood. J J Wood Res Soc 66(2):59–66 (in Japanese)
17. Yamashita O, Yokochi H, Miki T, Kanayama K (2009) The pliability of wood
and its application to molding. J Mater Proc Technol 209:5239–5244
18. Miki T, Seki M, Sugimoto H, Shigematsu I, Kanayama K (2013) Preparation
of three dimensional products using flow deformability of wood treated
by small molecular resins. Adv Mater Res 856:79–86
19. Seki M, Tanaka S, Miki T, Shigematsu I, Kanayama K (2017) Forward extru-
sion of bulk wood containing polymethylmethacrylate: effect of polymer
content and die angle on the flow characteristics. J Mater Proc Technol
239:140–146
20. Seki M, Tanaka S, Miki T, Shigematsu I, Kanayama K (2016) Extrudability
of solid wood by acetylation and in-situ polymerization of methyl meth-
acrylate. BioResources 11(2):4205–4036
21. Seki M, Kiryu T, Miki T, Tanaka S, Shigematsu I, Kanayama K (2016)
Extrusion of solid wood impregnated with phenol formaldehyde (PF)
resin: effect of resin content and moisture content on extrudability and
mechanical properties of extrudate. BioResources 11(3):7697–7709
22. Yin X, Huang A, Zhang S, Liu R, Ma F (2018) Identification of three Dalber-
gia species based on differences in extractive components. Molecules
23(9):1–11
23. Eyton WB, Ollis WD, Sutherland IO (1967) The neoflavanoids group of
natural products-I: Dalbergiones a new class of quinones. Tetrahedron
21(9):2683–2696
24. Donnelly BJ, Donnelly DMX, O’Sullivan AM, Predergast JP (1969) Dalber-
gia species VII, isolation and structure of melanoxin a new dihydroben-
zofuran from Dalbergia melanoxylon Guill. and Perr. (Leguminoseae).
Tetrahedron 25(18):4409–4414
25. Seshadri TR (1972) Polyphenols of Pterocarpus and Dalbergia woods.
Phytochemistry 11(3):881–898
26. Rasband WS (1997–2012) ImageJ, U. S. National Institutes of Health,
Bethesda, Maryland, USA. https ://rsb.info.nih.gov/ij/. Accessed 12 Mar
2020
27. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25
years of image analysis. Nat Methods 9:671–675
28. Shams MI, Yano H, Endou K (2006) Compressive deformation of wood
impregnated with low molecular weight phenol formaldehyde (PF)
resin I: effects of pressing pressure and pressure holding. J Wood Sci
50:337–342
29. Nzokou P, Kamdem DP (2004) Influence of wood extractives on moisture
sorption and wettability of red oak (Quercus rubra), black cherry (Prunus
serotina), and red pine (Pinus resinosa). Wood Fiber Sci 36(4):483–492
30. Hernández RE (2007) Swelling properties of hardwoods as affected by
their extraneous substances, wood density, and interlocked grain. Wood
Fiber Sci 39(1):146–158
31. Hashemi H, Latibari J (2011) Evaluation and identification of walnut heart-
wood extractives for protection poplar wood. BioResources 6(1):56–69
32. Mantanis GI, Young RA, Rowell RM (1994) Swelling of wood. Part I: swell-
ing in water. Wood Sci Technol 28(2):119–134
33. Mantanis GI, Young RA, Rowell RM (1995) Swelling of wood. Part III: effect
of temperature and extractives on rate of maximum swelling. Holzfor
49(3):239–248
34. Adamopoulos S, Voulgaridis E (2012) Effect of hot-water extractives on
water sorption and dimensional changes of black locust wood. Wood Res
57(1):69–78
35. Jankowska A, Drożdżek M, Sarnowski P, Horodeński J (2017) Effect of
extractives on the equilibrium moisture content and shrinkage of
selected tropical wood species. BioResources 12(1):597–607
36. Takamura N (1968) Studies on hot pressing and drying process in the
production of fiber board. III. Softening of fiber components in hot press-
ing of fiber mat. J J Wood Res Soc 14(2):75–79 (in Japanese)
37. Furuta Y, Nakajima K, Nakatani T, Kojiro K, Nakamaru Y (2008) Effects of
lignin on the thermal-softening properties of water-swollen wood. J Soc
Mater Sci 57:344–349 (in Japanese)
38. Furuta Y, Okuyama T, Kojiro K, Miyoshi Y, Kiryu T (2014) Temperature
dependence of the dynamic viscoelasticity of bases of Japanese cypress
branches and the trunk close to the branches saturated with water. J
Wood Sci 60:249–254
39. Matsunaga M, Minato K (1998) Physical and mechanical properties
required for violin bow materials II: comparison of the processing
properties and durability between Pernambuco and substitutable wood
species. J Wood Sci 44:142–146
40. McNaught AD, Wilkinson A (eds) (2014) Compendium of chemical ter-
minology, 2nd edn. Blackwell Scientific Publications, London. https ://doi.
org/10.1351/goldb ook
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub-
lished maps and institutional affiliations.
