See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/371279358

Through-thickness distribution of bamboo tensile strength parallel to fibres

Article in SN Applied Sciences · June 2023

DOI: 10.1007/s42452-023-05392-x

CITATIONS READS

0 213

2 authors:

Layth S. Al-Rukaibawi György Károlyi

Budapest University of Technology and Economics Budapest University of Technology and Economics
8 PUBLICATIONS 43 CITATIONS 84 PUBLICATIONS 1,619 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Layth S. Al-Rukaibawi on 04 June 2023.

The user has requested enhancement of the downloaded file.

Research

Through‑thickness distribution of bamboo tensile strength parallel

to fibres
Layth S. Al‑Rukaibawi1 · György Károlyi2

Received: 7 March 2023 / Accepted: 22 May 2023

© The Author(s) 2023  OPEN

Abstract
A tensile test program of Moso bamboo samples was carried out in order to obtain information on the radial distribu-
tion of elastic and strength properties. In accordance with previous studies, the elastic modulus and the tensile strength
were found to increase outwards in the radial direction. This spatial dependence was compared with the distribution
of the fibre content of the layers and found excellent correlation suggesting that the fibres are the main load bearing
components. The fibre content was quantified using regression analysis in the radial direction as 0.1 (inner layer), 0.19
(middle-1 layer), 0.28 (middle-2 layer) and 0.44 (outer layer). The Rule of Mixture theory allowed the estimation of the
elastic modulus and the tensile strength of the main individual components, the fibres and the parenchyma matrix.

Article Highlights

• Radial distribution of bamboo fibres across the culm • The radial distribution of bamboo fibres was analysed
wall and stiffness properties obtained from tensile test in layers using image analysis software.
parallel to fibres were compared. • The tensile failure mechanisms of bamboo samples
were identified and found to depend on fibre content.

Keywords Moso bamboo · Natural gradient fibre · Tensile testing · Failure mechanisms · Rule of mixtures

1 Introduction reinforced polymer [4]. The culm, the part of bamboo

plant above ground, is a wood-like material that contains
Bamboo as a bio-material has gained significant interest in two major components, the internodes and the nodes [7]
a broad range of sectors including buildings [1, 2], renew- (see Fig. 1a). The part of the culm right above the ground
able energy and many others [3]. Their light weight and (of about 2.1 m height) has the largest diameter and wall
fast-growing feature (3–4 years) while retaining stiffness thickness and lowest average internodal length in compar-
and thermal properties comparable to those of structural ison to the upper segments of the bamboo culm [8–10].
wood products [4, 5] allow them to spread globally [6]. Several bamboo species including Phyllostachys pube-
Additionally, bamboo specific strength-to-weight ratio scens or P. edulis referred herein as ‘Moso bamboo’, in par-
can be compared to low-carbon steel and glass fibre ticular, have been researched extensively. Moso bamboo

* Layth S. Al‑Rukaibawi, [email protected]; György Károlyi, [email protected] | 1Department of Structural

Engineering, Budapest University of Technology and Economics, Műegyetem rkp. 3., Budapest 1111, Hungary. 2Institute of Nuclear
Techniques, Budapest University of Technology and Economics, Műegyetem rkp. 3., Budapest 1111, Hungary.

SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

Vol.:(0123456789)
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

Fig. 1  Processing overview of bamboo strips with a bottom part of culm, b bamboo cross-section wall, c strips cutting, d short bamboo
strips and e vascular bundles and parenchyma matrix. Microscopic image (e) is taken from Ref. [15]

has a ring-like cross-section with a natural functionally During the last three decades since 1995, several experi-
graded microstructure [10] (see Fig. 1b). This structure mental investigations sought to determine through-culm
is considered as a natural composite material consisting gradient fibre content of Moso bamboo with their related
of two major components: the vascular bundles and the stiffness and strength properties in tension parallel to the
parenchyma matrix. The culm wall is composed of spongy fibres. Nogata and Takahashi [16], and Amada et al. [17]
tissue-like material called parenchyma cells (matrix) investigated the radial distribution of bamboo fibres at
(around 52% gross volume fraction, as a representative different locations from the inside to the outside of the
value) [11], stiff sclerenchyma-cells (fibre bundles) (about culm cross-section wall. The axial stiffness and axial ten-
40%) and conductive tissues (tube of vessels) (about sile strength were found to be within the range of 5–25
8%) [9], see Fig. 1e. The vascular bundles are distributed GPa and 100–400 MPa, respectively. Huaqiang Yu et al.
densely in the outer region of the wall and sparsely in the [18] investigated the radial variation in tensile stiffness
inner region [10, 12]. It has been shown that the strength and strength at three height locations along the culm.
of the outer wall layers is higher than that of the inner lay- They observed that longitudinal stiffness and strength
ers [12] due to the radial increase of the amount of fibres had a curvilinear regression from the inner layer towards
from the inside to the outside of the culm wall, whilst the outer layer, the obtained values were within the range
parenchyma matrix content increases reversibly from the of 8.49–32.49 GPa and 115.94–328.15 MPa, respectively,
outer towards the inner part [10, 13, 14]. at the top height segment of the culm. Shao et al. [14]
The general manufacturing process starts with a stand- quantitatively analysed the variation of tensile properties
ardised cutting of bamboo strips from the middle part of at two scale levels: macro-level across the radial thickness
a ring like culm wall after the removal of the skin at the of the internodes and micro-level along the vascular bun-
outer part and the pith ring at the inner part. The resulting dles. At the macro-level, they observed a linear relation-
end-product is the engineered bamboo product (EBP) with ship between fibre volume fractions and tensile stiffness
cross-section dimensions of 20 mm width and 6 mm thick- and strength properties. Also, they reported the tensile
ness, see Fig. 1d. Even though EBP provides good perfor- strength and elastic modulus to be within the range of
mance characteristics for bio-composite design, utilising 4.5–20 GPa and 79.24–294.75 MPa, respectively. Li and
the outer layers of bamboo is more beneficial due to their Shen [19] studied the elastic modulus parallel to the
high stiffness and strength compared to other layers as it fibres at different locations along the bamboo culm. They
will be seen in the next sections. proposed empirical equations to correlate stiffness and

Vol:.(1234567890)
SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x Research

strength to fibre volume fractions. Liu et al. [20] applied modes B and D. Shao et al. [14] obtained three distinct
advanced optical imaging, namely, confocal laser scan- failures from samples at inner, middle and outer layers
ning microscopy to measure accurately the elastic modu- corresponding to bamboo fibres content. Samples from
lus, strength and volume fraction for the vascular bundles inner layers showed a brittle characteristic, indicating fail-
of bamboo strips across the culm wall. They reported the ure within the parenchyma matrix. Samples from middle
tensile strength and the elastic modulus of bamboo strips layers show uneven combined fracture surface of paren-
taken from the inner part of the culm to be 101.85 MPa chyma matrix and fibres. Outer layer samples showed a
and 6.65 GPa, respectively, and those at the outer layer single bamboo splitting. The same fracture pattern was
to be 294.26 MPa and 30.3 GPa, respectively. Long et al. also investigated by Liu et al. [20], where samples from the
[21] studied the dependence of the elastic modulus and higher fibre content outer layer predominantly fractured
tensile strength parallel to the fibres on the radial direction at parenchyma matrix and fibres surfaces, whereas sam-
within the culm. They reported the tensile strength and ples with smaller fibre content from the inner layer failed
the elastic modulus of bamboo strips taken from the inner within the parenchyma matrix.
part of the culm to be 37.5 MPa and 6.87 GPa, respectively, In the literature cited above, there has been little quan-
and those at the outer layer to be 262.26 MPa and 27.34 titative analysis to understand the relative importance of
GPa, respectively. Recently, Akinbade et al. [22] conducted the non-uniform distribution of bamboo fibres through
a modified flat-ring flexural test to obtain the stiffness the thickness of the culm wall. In order to consider the
properties in tension parallel to the fibres. Additionally, distribution pattern effect of bamboo vascular bundles
they analysed microscopically the culm radial wall tissue. on the performance of the culm itself, it is beneficial to
The dependence of stiffness on moisture content was also divide the culm wall into many parts or thin layers in the
investigated by Amaral et al. [23]. radial direction. Therefore, there are two primary aims of
To conclude, the above investigations were focused this study:
on the tensile strength and elastic modulus parallel and
perpendicular to the fibres. The longitudinal behaviour 1. To critically examine a through-thickness distribution
is combined with the assumption of uniform distribution of bamboo fibre volume fraction and to investigate the
of bamboo fibres through the thickness of the culm wall axial strength and elastic stiffness properties parallel
is typically modelled as a fibre reinforced polymer (FRP) to the fibres. This is achieved by physically testing thin
material following the polymeric composite materials layers of about 1 mm average thickness at four differ-
analogy of randomly oriented long fibres. Hence, the rule ent layers from the inner towards the outer positions
of mixture (RoM) approach can be considered applicable of the bamboo culm wall.
to predict the longitudinal stiffness and strength prop- 2. To ascertain the variation of the distribution pattern
erties. However, recent evidence by [24] suggested that of bamboo vascular bundles in correlation to the
RoM works well in the axial direction, but seems not to mechanical performance of each layer through the
be able to capture adequately the bamboo behaviour radial direction. This objective can be attained by
in transversal directions as it cannot easily be measured analysing microscopically the thin layers using image
in experiment [25]. Also, they highlighted the need to analysis software to count the gradient fibre content
extensively modelling the orthotropic behaviour of bam- at each layer. This paper shows that the non-uniform
boo fibre and parenchyma matrix. This observation was radial distribution of fibres is in correlation to the vari-
already highlighted in the authors previous research on ation of stiffness and strength properties of bamboo
an anatomy-based numerical modelling of bamboo micro- layers.
structure using Easy PBC, a plugin tool in ABAQUS, which 3. To determine a specific composite property by the
accurately predicted the anisotropic material properties Rule of Mixture (RoM) theory, an analytical homogeni-
of Moso bamboo [26]. sation analysis. This is verified by the validity of Eqs. (9)
The failure mechanisms and modes of bamboo in ten- and (10), to estimate the elastic stiffness and strength
sion parallel to fibres have been documented and inves- of Moso bamboo constituents parallel to the fibres.
tigated. Richard and Harries [27] reported that bamboo
tensile test specimens exhibited six groups of failure The remainder of this paper is divided into the fol-
modes: Mode A: specimen failure within the tabbed grip lowing sections. Section 2 gives a brief overview of the
length, Mode B: failures at the interface of the grip and materials used and the coupon preparation for tensile
gauge length sample, Mode C: a tensile rupture across test parallel to bamboo fibres. Then, the measurement
the gauge length of specimen, Mode D: a single bamboo of oven-dry density and moisture content of the speci-
splitting, Mode E: a "brooming" failure which affects the mens are briefly described. Experimental uniaxial tensile
entire cross section, and Mode F: a combined failure of test results and measured bamboo fibre volume fractions

Vol.:(0123456789)
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

are also reported in Sect. 2. The measured elastic prop- rounds of carbonisation and air drying. The final strips had
erties, including stiffness modulus and tensile strength, a dimension of 2.1 m length, 30–40 mm arched width and
are compared to previous experimental results, and the 10 mm thickness. Finally, the strips were packaged and
failure mechanisms of bamboo samples were presented shipped to Budapest, Hungary.
in Sect. 3. Section 4 presents the analytical homogenisa- For tensile specimen preparation, 15 bamboo strips of
tion analysis using RoM to estimate the elastic stiffness 200 mm length were taken from the internodal region par-
and strength for Moso bamboo parallel to fibres. Finally, allel to the fibres (longitudinal axis x). The peeling process
the main conclusions and areas for further research are starts with the removal of the outer skin and inner pith
identified in Sect. 5. surfaces, see Fig. 2a. Strips are then machined on each side
to achieve a rectangular section. The rectangular strips are
then machined to required thickness using a manual lay-
2 Experimental programme ering tool, in which the required thickness is achieved by
bolts adjusting the threaded gears in order to cut layers,
2.1 Materials and tensile specimen preparation see Fig. 2b. The original rectangular strips had 8 mm thick-
ness, after the removal of the outer and inner layers, the
The bamboo strips were commercially supplied and layer from the strong outer layer had about 1 mm thick-
obtained by Dasso Group, supplier of bamboo products. ness, the next layers of moderate strength named Mid-
The strips, taken from the lower section of four-year-old dle-2 and Middle-1 had about 1 mm thickness each, and
Moso bamboo (Phyllostachys pubescens). First, the bam- the softer remaining layer had 1 mm thickness, see Fig. 2a.
boo culms were harvested and collected from Fuyang The final strips had a length of 200 mm (along the lon-
bamboo plantation which is in North China’s subtropical gitudinal axis x), a width of 20 mm (along the tangential
zone in Zhejiang province. The culm diameter at chest axis z), and a thickness of 1 mm (along the radial axis y),
height was between 100 and 120 mm and the average see Table 1.
cross-section thickness of the culm wall was about 10 mm. Coupon tension specimens were categorised into four
A cylindrical segment of the culm from the ground level groups: specimens from the outer layer were denoted as
up to 2.1 m was cut. Then, the culm segments of 2.1 m Bo, those from the inner layer as Bi whereas the middle lay-
length were pushed onto a spike machine, which had 8 ers are BM1 and BM2 for specimens from the middle-1 and
sharp fins around it, hence the bamboo was split into 8 middle-2 layers, respectively. Each group of the uniaxial
wedges with an arc of 45 degrees each. Then, to stabilise tensile coupon specimens were manufactured from 4 tabs
the moisture content of these strips between 8 and 12%, glued to a single bamboo strip, two on each end. The tabs
they were subjected to a preservation process with two are acrylic clamping sheets made of glass fibre composite

Fig. 2  Steps of manufacturing tensile specimens with a bamboo strip selection, b bamboo strip layering, c tensile coupon specimen prepa-
ration and, d 75 tensile coupon specimens (out of which 60 was tested)

Vol:.(1234567890)
SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x Research

Table 1  Geometrical Property Length Width Bamboo Tabs thick- Effective Tabs length
dimensions of the uniaxial thickness ness length
tensile specimen (unit: mm)
L bo t1 t2 l1 l2

Tensile specimen 200 20 1 5 100 50

materials, see Fig. 2c. The tabs of size 1050 mm × 2050 mm samples, one from each group, as listed in Table 3. Follow-
and 5 mm thickness are Üveto-I sheets supplied by the ing the national bamboo standard ISO 22157: 2019 [28],
Hungarian company R-VILL PLUSZ Kft. The epoxy adhe- the samples were placed in the oven at a temperature of
sive is Araldite 2011, a two-component paste of 300 g sup- 103 ℃ for 24 h, recording the mass at regular intervals.
plied from the Hungarian company Neosil Kft. The tabs The equilibrium moisture content (EMC) can be calculated
prepared for the test specimens were of 50 mm length, from the formula
20 mm width and 5 mm thickness, see Table 1. Thus, the m0 − mdry
specimens had an affective length of 100 mm. All dimen- EMC(%) = × 100 (2)
sions were measured using Caliper, a precision device hav- m0
ing an accuracy of at least 0.1 mm. The tension specimens’
dimensions shown in Table 2 were prepared conforming 2.3 Tension test parallel to fibres
to the national bamboo standard ISO 22157: 2019 [28].
Coupon tension specimens were categorised into four
2.2 Oven‑dry density analysis groups: specimens from the outer layer were denoted
as Bo, those from the inner layer as Bi whereas the mid-
In order to determine the density of the uniaxial tensile dle layers are BM1 and BM2 for specimens from the mid-
specimens, the dimensions were measured of four sam- dle-1 and middle-2 layers, respectively. For each group,
ples chosen at four layers at a precision of at least 0.1 mm 15 specimens were tested. All tension test specimens
to calculate their volume V0 following the national bam- were prepared according to the procedure described in
boo standard ISO 22157: 2019 [28]. Then, to measure the the national bamboo standard ISO 22157: 2019 [28]. As
dry density 𝜌 , the specimens were dried out to a con- described above, the specimens had dimensions 200 mm
stant oven-dry mass, and the weighing were carried out length (axis x), 20 mm width (axis z) and 1 mm thickness
immediately afterward. The mass mdry of the oven-dry (axis y). The gauge (effective) length l1 was 100 mm, as
test specimens were determined to a precision of 0.5% of shown in Fig. 3b. All test specimens were conditioned and
the original specimen mass m0. The drying is considered stored at a climate-room environment at a temperature
complete when the difference between subsequent mass of 20 ± 2 ℃ and relative humidity of 65 ± 5% for 24 h. The
measurements (mi) do not exceed 0.5% of the pre-drying moisture content EMC of the samples was measured to be
mass mo, see Table 3. The dry density 𝜌 can be obtained approximately 8% in average.
from the formula: For the tensile test, the German universal material test-
mdry ing machine Zwick Z250, equipped with 20 KN capacity
𝜌= (1) load head was applied, see Fig. 3a. Prior to commencing
Vo
the test, the specimen was placed inside the two tensile
The equilibrium moisture content of the uniaxial tensile grip jaws and the gauge area was marked with 4 points
specimen was determined after the tension tests for four using a glow marker. Once the positions had been fixed, a

Table 2  Physical determination Bamboo Length Width Thickness Volume Mass (g) EMC Density
of the oven-dry density and samples (mm) (mm) (mm) (mm3) % (g/cm3)
moisture content for tensile 0h 3h 2h
specimens
L bo t1 Vo mo mdry mi 𝜌

7 Bi 202 20.6 0.95 3953.14 1.967 1.826 1.826 7.72 0.46

3 BM1 201 20.5 0.94 3892.54 1.737 1.604 1.604 8.29 0.41
6 BM2 200 20.4 1.03 4244.42 2.263 2.088 2.088 8.38 0.49
5 BO 200.3 20.37 0.99 4039.51 1.982 1.846 1.846 7.36 0.45

Vol.:(0123456789)
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

Table 3  Statistical analysis of the radial distribution of Moso bamboo strength and stiffness
Coupon specimen n Statistic indicator Fibre volume Bamboo thickness Max. Load Tensile strength Modulus
fraction of elastic-
ity
(V f ) (%) (t 1 ) (mm) (F ) N (TS ) MPa (EL) GPa

Outer layer (Bo) 15 Mean 0.44 0.90 4503 250 22.8

STD – 0.12 303 29.6 3.17
COV – 13% 6.7% 12% 14%
Middle layer (BM2) 15 Mean 0.28 0.943 3389 176 13.3
STD – 0.063 178 10.6 1.24
COV – 6.6% 5.3% 6.1% 9.3%
Middle layer (BM1) 15 Mean 0.19 0.876 2489 140 9.50
STD – 0.13 256 16.2 0.70
COV – 15% 10% 12% 7.2%
Inner layer (Bi) 15 Mean 0.10 1.182 1924 84.2 6.29
STD – 0.36 273 19.6 0.65
COV – 31% 14% 23% 10%

Fig. 3  Schematic overview of the DIC system and the tension test specimen. a DIC system, b Coupon tension test specimen and c Sche-
matic representation of tension stress–strain curves

digital image correlation (DIC) system including a camera tension stress–strain curves of the four different specimen
(Mercury Monet, Sobriety, Kurim, Czech Republic assisted groups. The elastic modulus was calculated as
with LED lights) and a computer desktop were set up to ( )
monitor the field view of the gauge area and to record l1 F40 − F10
EL = ), (3)
the data in form of individual images during the test. The
(
bo ⋅t1 Δ40 − Δ10
DIC was used as the optical strain measurement system,
its resolution and acquisition frequency are 5 Megapixel where l1 is the gauge length,F40 is 40% of the ultimate load,
and 60 Hz, respectively. The distance of the DIC system and F10 is 10% of the ultimate load, Δ40 is the displacement
the camera from the testing machine apparatus was about at load F40 , Δ10 is the displacement at load F10 , bo is the
500 mm. The specimens were loaded at a constant dis- width and t1 is the thickness of the specimen. The tensile
placement rate of 1 mm/min up to fracture when tensile strength, also known as the modulus of rupture, was also
strain reached less than 2%. Figure 3c shows the schematic calculated as

Vol:.(1234567890)
SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x Research

F Af
TS =
bo ⋅t1
, (4) Vf [%] = × 100 (5)
Ac
where F is the ultimate load.

2.4 Image processing analysis

3 Results and discussion
This paragraph describes the procedure to determine the
distribution pattern of bamboo vascular bundles based on 3.1 Oven‑dry density analysis
Liu et al. [20], also explained in [13]. The distribution pat-
tern of bamboo vascular bundles was determined at each The results of the physical determination of the oven-dry
thin strip and the fibre volume fraction Vf along the radial density and moisture content for tensile specimens along
direction. This is achieved by polishing the surface of the the fibre direction are summarised in Table 2. The moisture
samples and then taking photographs of the transverse content EMC of the samples was measured to be approxi-
sections by a compact digital camera Canon SX432 IS in mately 8% in average.
a room with ceiling light. The resolution of the camera is
20 Megapixel with 45 × optical zoom, and the distance of 3.2 Fibre volume fraction measurements
images was about 21.60 cm. Eight thin slices, excluding the
pith part at the inner side and the skin at the outer side of The correlation between the fibre volume fractions Vf and
the culm, were photographed. The fibre volume fraction the normalised distance was analysed, which is the radial
Vf at each thin slice was measured by analysing the images distance of the layer from the innermost point of the culm
with the imageJ/FIJI processing software, see Fig. 4a–c. The wall divided by the thickness of the culm wall (that is, it is
digital images were converted to binary format to obtain 0 inside and 1 outside). Figure 5 shows the resulting cor-
black and white images. This is shown in Fig. 4c, with black relation, we see a continuous increase of Vf along the radial
background and white fibre vascular bundles. Then fibre direction from the innermost to the outermost regions. A
surface areas Af were traced and measured as well as the second order regression polynomial function was found
total surface area Ac of each thin slice. The volume fraction to describe the volume fraction Vf as a function of the nor-
Vf at each thin slice was calculated as malised radial distance from the innermost layer with a fit

Fig. 4  Image processing analysis of the radial distribution of Moso bamboo fibre volume fractions

Vol.:(0123456789)
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

inner (Bi), middle-1 (BM1), middle-2 (BM2) and outer (Bo) lay-
ers of the culm. From the graphs, one can see a clear trend:
the ultimate tensile load increases monotonically along
the radial direction of the bamboo cross-section with the
largest ultimate loads in the outer, and with the smallest
ultimate loads in the inner layer. The load–displacement
relationship is remarkably linear practically up to the sig-
nificant peak at the ultimate load (fu), the load is propor-
tional to the displacement almost along the whole tension
test. Finally, a stage of brittle fracture is reached with a rela-
tively small fluctuation of the ultimate load. At this load,
Fig. 5  Distribution of fibre volume fractions of Moso bamboo from the load–displacement curves exhibit a drastic reduction
the innermost to outermost regions in tensile load reaching zero. Despite the marked differ-
ences in the break force across the four investigated layers,
interestingly, there is quite small variation in the ultimate
­ 2 around 0.9825.
having a coefficient of determination R displacement at failure for all samples. The ultimate dis-
The regression function is placement values reached about 1.4 mm for the middle
layers BM1 and BM2, whereas they were around 1.6 mm for
Vf = 0.3144x 2 + 0.1916x + 0.0172 (6) Bi and Bo samples. The difference between the break force
where x is the normalised distance from the inner to the for the samples taken from the different layers is mainly
outer layer. The volume fraction Vf in the four main lay- attributed to the gradient density of bamboo fibres along
ers was as follows: Bi: 0.1, BM1: 0.19, BM2: 0.28 and Bo: 0.44. the radial direction. In particular, higher fibre density in
These findings are consistent with those of [18–22] and the outer layer results in higher tensile strength, whereas
support the authors’ previous research [26]. lower fibre density in the inner layer leads to lower tensile
strength. This finding matches those observed in earlier
studies [18–22] shown in Table 4.
3.3 Statistical analysis of tensile properties of Moso After obtaining the stiffness related parameters of the
bamboo samples samples, a regression analysis was used to compare the
gradient of the stiffness properties along the radial direc-
The results of the statistical analysis of the radial distri- tion of the bamboo cross-section with the variation of the
bution of Moso bamboo strength and stiffness in tension fibre volume fraction Vf. The tensile strength and elastic
along the fibre direction are summarised in Table 3. The modulus of Moso bamboo in the four layers are shown
coefficient of variation (COV.) at each thin coupon test in Fig. 7a and b, respectively. There was a significant posi-
group is equal to the ratio of the standard deviation (STD.) tive correlation between bamboo stiffness properties and
to the mean of n = 15 samples. It can be seen from Table 4 relative positions with a trend of continuously increasing
that the ultimate tensile loads change monotonically tensile strength and elastic modulus from the inner to the
along the radial direction of the bamboo cross-section, it is outer layers. The relative distance from the inner edge of
the lowest at Bi (inner layer) and gradually increased up to the ring-like cross-section to the investigated layers is 0.2,
the largest value at Bo (outer layer). The maximum tension 0.4, 0.6 and 0.8, see Fig. 7a and b. The outer layer Bo shows
loads at the four tested layers of bamboo were: Bi: 1924 N, significantly higher tensile strength than the other three
BM1: 2489 N, BM2: 3389 N and Bo: 4503 N. Also, the ratio of layers by the ratio of 3 for layer Bi, of 1.8 for BM1 and of 1.4
the maximum break force of Bo is 2.3 higher than that of for BM2. Similarly, the outer layer Bo showed significantly
Bi, 1.9 higher than that in BM1, and 1.3 higher than that in higher elastic modulus than the other three layers by the
BM2. It is worth mentioning that all layers’ properties have ratio of 3.6 for Bi, of 2.4 for BM1 and of 1.7 for BM2. These
shown a pronounced dispersion reflected in a high value observed trends are consistent with those found by [14,
of variation coefficient COV. This is caused by the variation 18, 20, 21] see Table 4, and in authors previous research
in material details of bamboo strips as a natural material about the anatomy based microstructure of Moso bam-
as well as their thickness. boo, see [26].
The regression analysis of tensile strength and mod-
3.4 Elastic modulus and tensile strength ulus with respect to the fibre volume fractions Vf is
shown in Fig. 7c and d, respectively. It can be observed
Figure 6 illustrates the load–displacement relationship in Fig. 7a and b that the increase of tension strength
curves during the stretching of bamboo strips from the and elastic modulus is correlated to the increase in fiber

Vol:.(1234567890)
 Table 4  Summary of gradient fibre volumes of Moso bamboo with their stiffness properties in tension, adopted from [22]
Refs. Methods Species B Gross section Inner Middle-1 Middle-2 Outer Proposed relationship
SN Applied Sciences

M (x = 0.2) (x = 0.4) (x = 0.6) (x = 0.8)

T
Vf EL Vf EL Vf EL Vf EL Vf EL
(GPa) (GPa) (GPa) (GPa) (GPa)

[16] Vf: IA P. edulis B – 9.1 0.09 2.5 – – – – 0.77 22.6 EL = ­(EL,x=0)e2.2x

EL: RoM T – 13.7 0.11 3.8 – – – – 0.88 33.8
(2023) 5:174

[17] Vf: IA P. edulis B 0.25 – 0.17 4 – – – – 0.5 29 Exponential ­Vf and EL

EL: tens M 0.28 – 0.18 6 – – – – 0.56 30
T 0.34 – 0.10 6 – – – – 0.6 32
[18] EL: tens P. pubescens B – – – 8.49 – 12.35 – 18.72 - 25 EL = 41.956x − 18.431
MOR = 334.941x − 93.847
M – – – 8.49 – 13.21 – 19.33 - 27.5
T – – – 8.49 – 15 – 25 - 32.49
[14] tens P. pubescens B – – 0.10 4.51 0.14 5.83 0.27 11.2 0.49 20 EL = 40.13x + 0.22
MOR = 562.69x + 19.04
[19] Vf: IA P. pubescens B – – 0.12 15.4 – – 0.45 23.14 0.67 28.3 Vf = 55x + 12.7
EL: tens. & RoM EL = 12.9x + 15.4
M – – 0.20 18 – – 0.49 25.56 0.69 30.6 Vf = 48.9x + 20.2
EL = 12.6x + 18
T – – 0.23 18 – – 0.55 27.06 0.77 33.1 Vf = 53x + 23.9
EL = 15.1x + 18
[20] Vf: CLSM P. pubescens B – – 0.08 6.65 0.12 8.93 0.21 14.2 0.47 30.3 Vf = 0.0442 ­e0.5676x
EL: tens EL = 57.82x + 0.165
MOR = 428.83x + 32.506
[21] Vf: IA P. pubescens – 0.26 – 0.11 6.87 0.14 8.16 0.22 14.5 0.42 27.34 EL = 40.839x + 0.448
EL: tens. & RoM
| https://doi.org/10.1007/s42452-023-05392-x

[22] Vf: P. edulis B 0.29 – 0.12 – – – – – 0.67 - Vf = 1.41x3 − 1.23x2 + 0.50x + 0.10
IA
This study Vf: IA P. pubescens B 0.30 – 0.10 6.28 0.19 9.54 0.28 13.3 0.44 22.8 Vf = 0.3144x2 + 0.1916x + 0.0172
EL: tens EL = 55.544x + 0.8257
MOR = 544.46x + 44.077

B = Bottom height of the culm; M = Middle height of the culm; T = Top height of the culm

Vol.:(0123456789)
Research
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

Fig. 6  Load–displacement curves from the tensile experiments of bamboo strips taken from various layers of the culm: a outer (Bo), b inner
(Bi), c middle-2 (BM2), and d middle-1 (BM1) and layer

volume fraction Vf, shown in Fig. 5. Based on this obser- 3.5 Typical failure mode mechanisms of bamboo
vation, one could expect a linear regression between samples
the stiffness properties and the fiber volume fraction,
which is validated by fitting straight lines in Fig. 7c and The ultimate load capacity of bamboo strips depends on
d. The obtained linear regression function for the ten- the failure mode and the amount of fibres present in the
sion strength TS, with a coefficient of determination ­R2 specific failure location. Now the tensile failure of bam-
around 0.8766 is boo in the various layers are examined. As outlined in the
national bamboo standard ISO 22157: 2019 [28], there are
TS = 544.46Vf + 44.077 (7) six general groups of tensile coupon failures. The tensile
The elastic modulus EL was also found to follow a lin- coupons of bamboo applied in this study exhibited five
ear regression function with respect to the fiber volume distinct mechanisms of failure, as depicted in Fig. 8 and
­ 2 around
fraction with a coefficient of determination of R Table 5. These failure mechanisms are as follows. Mode
0.9069 as A, which entails specimen failure within the tabbed grip
length, was not observed because strong glass fibre rein-
EL = 55.544Vf + 0.8257 (8) forced polymer tabs were applied rather than wooden
ones. Mode B pertains to failures at the interface of the
As it can be seen that the stiffness depends essen-
grip and sample. Mode C constitutes a tensile rupture
tially linearly on the amount of fibres, hence, the major
across the specimen. Modes D and E both involve longi-
part of the stiffness of the bamboo samples is attributed
tudinal splitting failures along the length of the specimen,
to the fibres with a relatively small part coming from the
with Mode D comprising a single bamboo splitting, and
stiffness of the matrix.

Vol:.(1234567890)
SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x Research

Fig. 7  Gradient stiffness properties along the radius of the culm tance from inner to outer locations, c tensile strength versus fibre
with a tensile strength parallel to fibres versus distance from inner volume fraction, and d elastic modulus versus fibre volume fraction
to outer locations, b elastic modulus parallel to fibres versus dis-

Fig. 8  Failure modes of bamboo samples parallel to fibres direction as observed in the experiments

Mode E refers to a "brooming" failure which affects the (42%) exhibited a mixed mode failure, mode F. This is a
entire cross section. Mode F represents a combination of composite failure consisting of a mode B rupture failure
failure modes B and D. and a mode D splitting failure. Seven specimens (11%)
From the 60 investigated tensile coupons (Table 5), exhibited a mode D longitudinal splitting failure alone.
regardless of the radial position of bamboo, 25 samples Fifteen specimens (26%) experienced failures in mode

Vol.:(0123456789)
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

Table 5  Representative Coupon specimen n Failure modes

failure modes of the radial
distribution of Moso bamboo B C D E F
slices
Outer layer (Bo) 15 0 3 3 1 8
Middle layer (BM2) 15 4 1 2 1 7
Middle layer (BM1) 15 2 4 2 1 6
Inner layer (Bi) 15 9 2 0 0 4

B rupture near the tab area. Ten specimens (16%) were

( )
TS = 544.46Vf + 44.077 = 588.53Vf + 44.077 1 − Vf
observed to have mode C tensile rupture, and only 3 (10)
specimens (5%) exhibited mode E fibre splitting explosion
That is, the tensile strength of fibre and paren-
at the mid-region. It was discovered that samples with a
chyma matrix are calculated as 𝜎f = 588.53 MPa and 𝜎p =
higher fibre content, that is, from the outer layer of the
44.077 MPa, respectively. These numerical values are also
culm, predominantly failed via mode F failure, whereas
consistent with those reported previously [20, 21].
samples with a lower fibre content, that is, from the inner
It can be found that the RoM theory and the measured
layer, failed within the parenchyma matrix, as evidenced
correlation between the stiffness and strength properties
by mode B failure. This observation is consistent with the
and the fibre volume fraction are consistent, and allow
findings reported by [27] and [20].
for a simple estimation of the properties of the individual
components (fibre and matrix).
4 Estimate of the elastic modulus
and strength for Moso bamboo parallel
5 Conclusions and future work
to fibres
This study critically examined the radial distribution of
In a previous study [26], the authors proposed an anat-
the stiffness properties of bamboo by dividing the culm
omy-based representative volume element (RVE) model
wall into thin strips at different radial layers and uniaxi-
utilising the EasyPBC homogenisation tool in ABAQUS
ally testing the specimens in tension. The tension test
FEA software [29] to estimate the stiffness properties
programme was performed on sixty coupon samples of
of the modelled gradient volume fractions of RVEs. The
Moso bamboo (Phyllostachys pubescens) parallel to fibres.
vascular bundles and parenchyma matrix can be homog-
The samples were prepared by dividing the bamboo culm
enized into a fibre/matrix region, whose properties can be
wall into thin layers of about 1 mm average thickness at
conveniently determined by the implementation of the
four different positions from the inner towards the outer
well-established rule of mixture (RoM) theory. Based on
layers. The main aim of this study was to critically examine
the linear Eq. (8) and [18] established between the elastic
a through-thickness distribution of stiffness and strength
modulus EL and the fibre volume fraction Vf of the bamboo
properties of bamboo (like elastic modulus and axial ten-
samples, it is found
sile strength). From the experimental programme, the
elastic stiffness and strength properties can be obtained
( )
EL = 55.544Vf + 0.8257 = 56.369Vf + 0.8257 1 − Vf (9)
in four layers of the culm. In addition, the oven-dry density
from this, the elastic modulus of the fibre Ef and paren- and moisture content of the specimens were determined.
chyma matrix Ep can be approximated as Ef = 56.369 GPa The relation between the distribution of the vascular bun-
and Ep = 0.825 GPa. These values are consistent with those dles and the mechanical performance of each layer were
reported elsewhere [14, 18, 20, 21]. It is evident that the investigated. The layers have been analysed microscopi-
fibre serves as the primary load-bearing component, cally using image analysis software to count the distribu-
which exerts a significant influence on the tensile prop- tion of fibre content at each layer. The failure mechanisms
erties of bamboo. Additionally, the findings lend support of bamboo samples were further discussed and presented.
to the hypothesis that bamboo is a unidirectional fibre- Finally, an analytical homogenisation analysis, using rule of
reinforced composite material. mixture theory was reviewed to estimate the elastic stiff-
Analogously, a correlation (7) has been established ness and strength of Moso bamboo constituents parallel
between the measured tensile strength TS and the fibre to the fibres.
volume fraction Vf of the bamboo samples. Applying the Based on the results obtained from this study, the fol-
RoM theory, the formula can be written as follows: lowing can be concluded:

Vol:.(1234567890)
SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x Research

1. The tension strength in the four layers are correlated to Author contributions LSAR: Conceptualization, Methodology, Writ-
their radial positions from the inner towards the outer ing—original draft, Physical testing, Formal analysis, Validation,
Resources. GK: Writing—review and editing, Supervision, Funding
layer as follows: Bi: 84 MPa, BM1: 140 MPa, BM2: 181 MPa, acquisition.
and Bo: 249 MPa. This implies that the outer layer Bo
has significantly higher tensile strength than the other Funding Open access funding provided by Budapest University of
three layers by the ratio of about 3.0 at Bi, of 1.8 at BM1, Technology and Economics. This work was supported by the Hun-
and of 1.4 at BM2.The elastic modulus EL , in the four garian NKFIH under grant No. K-128584, and by the NRDI Fund of
Hungary under grant TKP2021-NVA-02. L S A-R was supported by
layers are also correlated to their radial positions from the Stipendium Hungaricum scholarship scheme.
the inner towards the outer layer as follows: Bi: 6.4 GPa,
BM1: 9.6 GPa, BM2: 13.3 GPa, and Bo: 22.8 GPa. The outer Data availability All data that support the findings of this study will
layer Bo has significantly higher elastic modulus than be available from the authors upon request.
that of the other three layers by the ratio of about 3.6
at Bi, of 2.4 at BM1, and of 1.7 at BM2. Declarations
2. As a result of image analysis of bamboo cross-sec-
Conflict of interest The authors declare that they have no compet-
tions, distribution of fibre volume fraction Vf in the ing financial interests or personal relationships that could have ap-
radial direction of the culm wall was determined and peared to influence the work reported in this paper.
described by a second order regression polynomial
function. The fibre volume fractions in the radial direc- Open Access This article is licensed under a Creative Commons Attri-
tion varied as follows: Bi: 0.1, BM1: 0.19, BM2: 0.28, and bution 4.0 International License, which permits use, sharing, adap-
tation, distribution and reproduction in any medium or format, as
Bo: 0.44. long as you give appropriate credit to the original author(s) and the
3. It was observed that samples with a higher fibre con- source, provide a link to the Creative Commons licence, and indicate
tent predominantly failed via mode F failure (com- if changes were made. The images or other third party material in this
bined failure due to failure at grip and delamination article are included in the article’s Creative Commons licence, unless
indicated otherwise in a credit line to the material. If material is not
of a single fibre), whereas samples with a lower fibre included in the article’s Creative Commons licence and your intended
content by mode B failure (failure near grip). use is not permitted by statutory regulation or exceeds the permitted
4. It is found that the application of the Rule of Mixture use, you will need to obtain permission directly from the copyright
theory was applicable to estimate the tensile strength holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​
org/​licen​ses/​by/4.​0/.
and the elastic modulus of the individual components
of bamboo, that is, of the fibres and the parenchyma
matrix, from our experimental data. This was achieved
by comparing the measured dependence of the elastic
References
modulus and the axial strength on the radial direction
to that of the fibre volume fraction using regression 1. Zhao J-C, Qiu H-X (2023) Seismic performance assessment of a
analysis. multi-story bamboo frame structure. Adv Bamboo Sci 2:100011.
https://​doi.​org/​10.​1016/j.​bamboo.​2022.​100011
We note that the similarity of the failure modes 2. Faridul Hasan KM, Noman Al Hasan KM, Ahmed T, György S-T,
Pervez MN, Bejó L, Sándor B, Alpár T (2023) Sustainable bamboo
observed in our experiments are very similar to those fiber reinforced polymeric composites for structural applica-
observed for polymer and composite materials. This leads tions: A mini review of recent advances and future prospects.
to the conjecture that bamboo can be looked at as a natu- Case Stud Chem Environ Eng 8:100362. https://​doi.​org/​10.​
ral fibre reinforced polymer matrix composite. 1016/j.​cscee.​2023.​100362
3. Chang C-W, Lin C-Y, Kao H-W, Chang F-C (2023) Micromechanical
As a possible next step, the authors expect to develop analytical and inverse numerical modeling methods for deter-
a reliability-based analysis of outer layer bamboo for wind mining the transverse elastic modulus of bamboo. Wood Sci
turbine applications to assess their mechanical reliability Technol 57:75–92. https://d ​ oi.o
​ rg/1
​ 0.1
​ 007/s​ 00226-0 ​ 22-0
​ 1445-3
considering a wide range of uncertainties at the laminate 4. Mahdavi M, Clouston PL, Arwade SR (2012) A low-technology
approach toward fabrication of Laminated Bamboo Lumber.
scale including lamina ply thickness (t), ply orientation Constr Build Mater 29:257–262. https://d ​ oi.o
​ rg/1
​ 0.1
​ 016/j.c​ onbu​
(theta) and loadings. Another interesting direction could ildmat.​2011.​10.​046
be to correlate not only the fibre density but also the fibre 5. Wu M, Mei L, Guo N, Ren J, Zhang Y, Zhao Y (2022) Mechani-
thickness to the strength. In general, thinner fibres can be cal properties and failure mechanisms of engineering bamboo
scrimber. Constr Build Mater 344:128082. https://​doi.​org/​10.​
associated with higher mechanical performance, but its 1016/j.​conbu​ildmat.​2022.​128082
precise role in bamboo is to be investigated further. 6. Farrelly D (1996) The book of bamboo: a comprehensive guide
to this remarkable plant, its uses, and its history. Thames and
Hudson Ltd, London

Vol.:(0123456789)
Research SN Applied Sciences (2023) 5:174 | https://doi.org/10.1007/s42452-023-05392-x

7. Ahmad M, Kamke FA (2005) Analysis of Calcutta bamboo for 20. Liu H, Jiang Z, Zhang X, Liu X, Sun Z (2014) Effect of fiber on
structural composite materials: physical and mechanical prop- tensile properties of moso bamboo. BioResources 9:6888–6898.
erties. Wood Sci Technol 39:448–459. https://​doi.​org/​10.​1007/​ https://​doi.​org/​10.​15376/​biores.​9.4.​6888-​6898
s00226-​005-​0016-y 21. Long L, Wang Z, Chen K (2015) Analysis of the hollow structure
8. Amada S, Untao S (2001) Fracture properties of bamboo. Com- with functionally gradient materials of moso bamboo. J Wood
pos Part B Eng 32:451–459. https://​doi.​org/​10.​1016/​S1359-​ Sci 61:569–577. https://​doi.​org/​10.​1007/​s10086-​015-​1504-9
8368(01)​00022-1 22. Akinbade Y, Harries KA, Flower CV, Nettleship I, Papadopoulos
9. Liese W (1998) The anatomy of bamboo culms. Brill C, Platt S (2019) Through-culm wall mechanical behaviour of
10. Londoño X, Camayo GC, Riaño NM, López Y (2002) Characteriza- bamboo. Constr Build Mater 216:485–495. https://​doi.​org/​10.​
tion of the anatomy of Guadua angustifolia (Poaceae: Bambu- 1016/j.​conbu​ildmat.​2019.​04.​214
soideae) culms, Bamboo Sci. Cult J Am Bamboo Soc 16:18–31 23. do Amaral LM, Rodrigues CS, Poggiali FSJ (2023) Assessment
11. Yang X, Pang X, Liu X, Yang S, Li X (2022) Determining the pore of physical, mechanical, and chemical properties of Dendro-
structure and radial variability of moso bamboo (Phyllostachys calamus asper bamboo after application of wetting and drying
edulis). Wood Sci Technol 100:1–13. https://​doi.​org/​10.​1007/​ cycles. Adv Bamboo Sci 2:100014. https://​doi.​org/​10.​1016/j.​
s00226-​022-​01451-5 bamboo.​2022.​100014
12. Lo TY, Cui HZ, Leung HC (2004) The effect of fiber density on 24. Akinbade Y, Harries KA (2021) Is the rule of mixture appropri-
strength capacity of bamboo. Mater Lett 58:2595–2598 ate for assessing bamboo material properties? Constr Build
13. Huanrong L (2010) Study on the properties and mechanism of Mater 267:120955. https://d ​ oi.o
​ rg/1
​ 0.1
​ 016/j.c​ onbui​ ldmat.2
​ 020.​
fracture in bamboo. Ph.D. dissertation, China Knowledge Net- 120955
work (CNKI) 25. Akinbade Y, Nettleship I, Papadopoulos C, Harries KA (2021)
14. Shao ZP, Fang CH, Huang SX, Tian GL (2010) Tensile properties Modelling full-culm bamboo as a naturally varying function-
of Moso bamboo (Phyllostachys pubescens) and its components ally graded material. Wood Sci Technol 55:155–179. https://​doi.​
with respect to its fiber-reinforced composite structure. Wood org/​10.​1007/​s00226-​020-​01246-6
Sci Technol. https://​doi.​org/​10.​1007/​s00226-​009-​0290-1 26. Al-Rukaibawi LS, Omairey SL, Károlyi G (2021) A numerical anat-
15. Dixon PG, Gibson LJ (2014) Sample of the Data Generated in the omy-based modelling of bamboo microstructure. Constr Build
Structural Bamboo Project of the Cellular Solids Group (at MIT), Mater 308:125036. https://d ​ oi.o
​ rg/1
​ 0.1
​ 016/j.c​ onbui​ ldmat.2
​ 021.​
DSpace@MIT. https://​dspace.​mit.​edu/​handle/​1721.1/​88182. 125036
Accessed 20 Jan 2021 27. Richard MJ, Harries KA (2015) On inherent bending in tension
16. Nogata F, Takahashi H (1995) Intelligent functionally graded tests of bamboo. Wood Sci Technol 49:99–119. https://​doi.​org/​
material: bamboo. Compos Eng 5:743–751 10.​1007/​s00226-​014-​0681-9
17. Amada S, Munekata T, Nagase Y, Ichikawa Y, Kirigai A, Zhifei Y 28. ISO 22157 (2019) Bamboo structures—Determination of physi-
(1996) The mechanical structures of bamboos in viewpoint of cal and mechanical properties of bamboo culms—Test meth-
functionally gradient and composite materials. J Compos Mater ods, Switzerland, Int. Organ. Stand.
30:800–819. https://​doi.​org/​10.​1177/​00219​98396​03000​703 29. Omairey SL, Dunning PD, Sriramula S (2019) Development of an
18. Yu H, Fei B, Ren H, Jiang Z, Liu X (2008) Variation in tensile prop- ABAQUS plugin tool for periodic RVE homogenisation. Eng Com-
erties and relationship between tensile properties and air-dried put 35:567–577. https://​doi.​org/​10.​1007/​s00366-​018-​0616-4
density for moso bamboo. Front For China 3:127–130. https://​
doi.​org/​10.​1007/​s11461-​008-​0008-9 Publisher’s Note Springer Nature remains neutral with regard to
19. Li H, Shen S (2011) The mechanical properties of bamboo and jurisdictional claims in published maps and institutional affiliations.
vascular bundles. J Mater Res 26:2749–2756. https://​doi.​org/​10.​
1557/​jmr.​2011.​314

Vol:.(1234567890)

View publication stats