the influence of humidity on mechanical properties of...

Construction and Building Materials 150 (2017) 35–48

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

The influence of humidity on mechanical properties of bamboo forbicycles

http://dx.doi.org/10.1016/j.conbuildmat.2017.05.1890950-0618/� 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: [email protected] (S. Jakovljevic).

Suzana Jakovljevic ⇑, Dragutin Lisjak, Zeljko Alar, Frano PenavaDepartment for Material, Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, Ivana Lucica 5, 10002 Zagreb, Croatia

h i g h l i g h t s

� Mechanical properties of two different bamboos in dry (environmental) and wet condition.� The change of moisture content in specimens is min 0,39% and max. 0,73%.� Both bamboos have lower values of mechanical properties after 3 weeks in wet chamber.

a r t i c l e i n f o

Article history:Received 30 October 2016Received in revised form 14 March 2017Accepted 22 May 2017Available online 1 June 2017

Keywords:BambooHumidityTensile strengthBending strengthCompressive strength

a b s t r a c t

Bamboo as an eco-friendly material has potential as construction material and proves his suitability inuse as an alternative for more traditional materials such as steel, aluminium and composite. This workpresents the influence of humidity on mechanical properties of two different bamboos, Pseudosasa ama-bilis (or Tonkin Cane) and Pleioblastus amarus (or Ku Zhu) which are used for bicycle frames. A large num-ber of tensile, compression and bending tests were carried out in two conditions, dry (in terms of theenvironment) and wet (after 3 weeks in wet chamber). It is shown that both, Tonkin Cane and Ku Zhu,in spite of low change in weight after 3 weeks in wet chamber have lower values of mechanical proper-ties. The results show that the tensile, compressive and bending strength of bamboo significantlydecreased after bamboo samples were kept in an environment with humidity level of 60%. The reductionof strength in the wet state for both types of bamboo is statistically proven by two sample comparisons oftheir arithmetic means at significance level a = 0.05. The results of this study indicate that the minimumchange of moisture in samples decreases bending, tensile and compression strength of bamboo. Potentialapplication of the results of this research is in the design and production of bicycle frameworks.

� 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Bamboo is natural construction composite material and thereare over 1250 different botanical types of bamboo in the world.It grows very fast and it can be harvested after 3 years, dependingon the species [1]. Therefore, global interest and research of its useare growing in many different applications in engineering and incivil construction [2–4]. Most of the research is focused on bamboofiber–reinforced composites [5,6] or for constructional purpose[1,3]. Bamboo as an eco-friendly material has been used for fabri-cation of bicycles [7,8] as a cheap alternative to traditional materi-als like steel or aluminium alloys. It has high strength to weightratio [8] which is very important for the structural design.

There is a lot of research on the mechanical properties of bam-boo [9–19]. The mechanical properties of bamboo are determinedby properties of fibers and matrixes and fibre density [6,11,19].Some studies have researched the fracture and toughening mech-anisms in bamboo structures [20,21] and the effects of microstruc-ture on the micro mechanisms of crack growth [11]. These studieshave shown that bamboo is susceptible to inter-laminar fracture,but the influence of outside moisture on crack behaviour has notbeen quantified. Some of them, like work by Askarinejad et al.[22] have investigated the effects of humidity on shear behaviourof bamboo. In this work it is shown that the specimens kept inenvironment with humidity levels between 60% and 80% had thehighest shear modulus and drop after it. The results show thatbamboo exhibits more ductility under torsion as the humidity ofthe samples increases. The results of this study provide significantinformation on the role of water on the optimal shear strength of

http://crossmark.crossref.org/dialog/?doi=10.1016/j.conbuildmat.2017.05.189&domain=pdf

http://dx.doi.org/10.1016/j.conbuildmat.2017.05.189

mailto:[email protected]

http://dx.doi.org/10.1016/j.conbuildmat.2017.05.189

http://www.sciencedirect.com/science/journal/09500618

http://www.elsevier.com/locate/conbuildmat

Fig. 1. Scanning electron microscope images of Tonkin Cane bamboo. (a) cross-section of Tonkin cane bamboo. (b) vascular bundles of fibers.

Fig. 2. Scanning electron microscope images of Ku Zhu bamboo. (a) cross-section of Ku Zhu. (b) vascular bundles of fibers.

Table 1Samples testing plan.

Types of bamboo Number of samples

Tonkin Cane Ku Zhu

Tests/Condition Dry Wet Dry Wet

Tension test 5 6 5 5Compression test 5 5 5 5Bending test 5 5 5 5

36 S. Jakovljevic et al. / Construction and Building Materials 150 (2017) 35–48

bamboo and its influence on mechanical properties of bamboofibre and matrix.

There have been some prior studies of mechanical behaviour ofround form (culm) specimens [13,18]. These studies have shownthat chosen types of bamboos are acceptable construction materi-als with excellent mechanical properties. Mechanical properties ofBambusa Pervariabilis (or Kao Jue) and Phyllostachy Pubescens (orMao Jue) which are used in bamboo scaffoldings have been stud-ied in [18] in terms of compression and bending strength in bothdry and wet condition. The specimens in wet tests were

immersed in water over different time periods and tests showedthat mechanical properties decreased in bamboo exposed to highhumidity.

Some of the primary properties that are affected by moisturecontent include: dimension, weight, strength (tension, compres-sion, bending), etc. This paper presents the findings on influenceof humidity on mechanical properties (tension, compression andbending) of two bamboo species, Pseudosasa amabilis (hereinafter,Tonkin Cane) and Pleioblastus amarus (hereinafter, Ku Zhu) that areused for bamboo bicycle frames. The first part of paper investigatesthe mechanical properties of two types of bamboos in dry (envi-ronmental) condition and the second part the mechanical proper-ties after 3 weeks in wet chamber.

2. Materials and methods

Both bamboo (Tonkin Cane and Ku Zhu) materials were obtained from China.Every pole was cut from a bamboo culm of over 3 years of age (Tonkin cane wascut in 2012 and Ku Zhu in 2013). After harvest they were dried for six months ina warehouse with 30% humidity without any chemical treatment and then follow-ing laboratory conditions, they were stored in ventilated warehouse for at least ayear.

Fig. 3. Tension test (a) Specimen dimension. (b) Force-crosshead position curve for sample 2–5 Tonkin Cane (c) Specimen after testing.

Table 2The weight loss of specimens.

Specimen Weight before g Weight after g Dm g w (H2O) Change%

Tonkin Cane1 279,0000 280,9400 19400 0,006905 069053-A 251,0000 252,8540 18540 0,007332 073323-B 189,8315 191,1598 13283 0,006949 069499-A 92,5486 93,1733 06247 0,006705 067059-B 72,8660 73,2951 04291 0,005854 05854

Ku Zhu22-A 167,1008 168,1640 10632 0,006322 0632222-B 159,0261 159,9592 09331 0,005833 0583324-A 58,8118 59,0454 02336 0,003956 0395624-B 47,2508 47,4816 02306 0,004861 04861

S. Jakovljevic et al. / Construction and Building Materials 150 (2017) 35–48 37

2.1. Specimen preparation

The condition of bamboo culms was estimated by visual examination and theywere marked. The length, outer diameter and thickness of both Tonkin Cane and KuZhu were measured. Samples for each testing were cut randomly from bambooculms.

The length of Tonkin Cane and Ku Zhu was between 565 and 709 mm. Thediameters of Tonkin Cane were 38,3–62,2 mm and 22,5–26,2 mm and for Ku Zhu23,2–46,1 mm. The thickness ranged between 3,3–7,5 mm and between 4,4–8,5 mm, respectively. Comparing the thickness and diameters of both bamboosthere are no indications that the thickness of bamboo is proportional to the diam-eter of bamboo.

The microstructure of Tonkin Cane and Ku Zhu was characterized using scan-ning electron microscope (SEM) Tescan Vega 5136 mm. Prior to scanning the sam-ples, one of each type of bamboo was cut in transverse cross- section of bambooculm, placed in EMITECH sputter coater and deposited by palladium and gold for120 s for better conductivity. The surface morphologies of samples were analyzedby secondary electrons. The microstructures of both bamboos were taken perpen-

dicular to face shown on Figs. 1 and 2. This figures show bamboo structure consist-ing of vascular bundles of fibers in the matrix of lignin.

As shown in Figs. 1 and 2 the difference in the structure, shape and distributionof vascular fibers in Tonkin Cane and Ku Zhu bamboo is obvious. The imagesdemonstrate a structure that consists of individual fibers in vascular bundles.

Tension, compression and bending tests were carried out on Heckert testmachine, type WPM, EU 40 MOD, accuracy class 0.5. In order to determine influenceof humidity on mechanical properties of bamboo two test conditions were carriedout–dry tests and wet tests as follows:

(a) Dry tests. The tests were intended to measure the tension, compression andbending strengths of the test specimens in dry laboratory condition, at tem-perature of 22 �C. Tension, compression and bending tests were carried outon five specimens of Tonkin Cane and Ku Zhu bamboo.

(b) Wet tests. The tests aimed at measuring the tension, compression andbending strengths of the test specimens after being exposed in the wetchamber with relative humidity (60 ± 2)% and temperature (20 ± 2)�C for21 days. Since bamboo has hollow culms, ends of culms were covered with

Fig. 4. Compression test (a) Dimensions of bamboo specimen. (b) Force-crosshead position curve. (c) Failure mode – splitting.

Fig. 5. Bending test (a) Schematic of the bending test (b) Force-crosshead position curve. (c) Failure mode – splitting.


Table 3Results of static tensile test for Tonkin Cane.

No. Sample a0 mm b0 mm S0 mm2 Fm kN Rm N/mm2

Dry condition (Ts_dTC)1. 2-1 3,78 5,58 21,09 4645 220,222. 2-2 4,04 5,48 22,14 4766 215,273. 2-3 4,43 5,53 24,50 4465 182,264. 2-4 4,33 4,99 21,61 4064 188,095. 2-5 3,99 5,50 21,95 5402 246,16

Average 4,11 5,47 22,29 4668 210,40

Wet condition (Ts_wTC)1. 1-A-1 4,49 5,40 24,25 5414 223,292. 1-A-2 4,20 5,38 22,60 4854 214,823. 1-A-3 3,98 5,10 20,30 2092 103,064. 1-A-4 3,85 5,22 20,10 2296 114,255. 1-A-5 3,95 5,27 20,82 4884 234,626. 1-A-6 4,02 5,24 21,06 3072 145,84

Average 4,08 5,27 21,52 3768 172,65

Table 4Results of static tensile test for Ku Zhu.

No. Sample a0 mm b0 mm S0 mm Fm kN Rm N/mm2

Dry condition (Ts_dKZ)1. 18–1 4,35 5,35 23,27 4520 194,222. 18–2 4,27 5,35 22,84 4271 186,963. 18–3 4,67 5,30 24,75 4915 198,584. 18–4 4,48 5,38 24,10 4815 199,775. 18–5 4,66 5,20 24,23 4564 188,35

Average 4,49 5,32 23,84 4620 193,58

Wet condition (Ts_wKZ)1. 22-B-1 6,70 4,76 31,89 5327 167,032. 22-B-2 5,93 5,30 31,43 6260 199,183. 22-B-3 6,42 5,38 34,54 5070 146,794. 22-B-4 6,26 5,28 33,05 3758 113,705. 22-B-5 6,20 4,81 29,82 4474 150,02

Average 6,30 5,11 32,15 4979 155,34


Parafilm M(R) to simulate realistic environmental condition i.e. to allowmoisture to penetrate only through outer wall. After 21 days, stalks wereremoved, cut to specimen dimension and immediately tested. Tension,compression and bending tests were carried out on five specimens of Ton-kin Cane and six specimens of Ku Zhu bamboo. The samples testing plan indry and wet condition is shown in Table 1.

2.2. Tension

The tensile properties of bamboo have been measured in some researches[11,20,23]. Bamboo slices in this work were obtained along the longitudinal direc-tion and afterwards separated with equal space in the cross-section. Finally, speci-mens used for tensile experiment were cut from these bamboo slices as in Fig. 3a.Tests were performed according to norm HRN EN ISO 6892-1 at temperature of22 �C and loading rate of 10 mm/min. The specimens were loaded continuouslyuntil failure at a displacement rate of 10 mm/min until failure as shown in Fig. 3c.

2.3. Compression

Bamboo specimens for compression strength testing were cut from culms in thelength of 50 mm and the dimension was measured according to Fig. 4a. The ratio d1/d2 was kept less than 1.2 to exclude the effect of buckling failure during the com-pression test, d1 – external diameter, mm, d2 – internal diameter, mm, b – thicknessof wall of bamboo, mm. The bamboo was tested under compressive load betweensteel plates with friction-free surface as shown in Fig. 4d. The specimens wereloaded continuously until failure at a displacement rate of 5 mm/min. The compres-sive strength was determined by maximum force before the appearance of the firstsplitting of test samples.

2.4. Three point bending test

Samples for bending test were 200 mm in length with arbitrary diameters. Eachspecimen was supported at the distance L = 76 mm and tested under single point

load F at L/2 until failure as in Fig. 5a. Load was perpendicular to fibers in matrix.The specimens were loaded continuously until failure at a loading rate of 5 mm/min. The bending strength was determined by maximum force before the appear-ance of the first splitting of test samples.

2.5. Statistical test

In order to determine the differences between the strength of dry and wet sam-ples, the work was carried out comparing the means of two independent samplesand all three types of mechanical load. The performed statistical test comparingthe means of two independent samples was used to determine whether or not thereare significant differences between the means of the populations from which thesamples were taken. The differences are defined as D = m1 � m2 where m1 – meanof dry sample and m2 – mean of wet sample.

The t-test for testing the null hypothesis was also performed. The hypothesesthat were tested:

Null Hypothesis H0 : D ¼ 0Null Hypothesis H1 : D–0

Associated with each t-statistic is a p-value. Small p-values (less than 0.05 ifoperating at the 5% significance level) lead to rejection of the null hypothesis. Sum-mary statistics (measures of central tendency, measure of dispersion, measures ofshape) in the tables show a number of different statistics data that are commonlyused to summarize a sample of variable data. To illustrate the important featuresof numeric data the Box-and-Whisker diagram was designed. The Box-and-Whisker plot summarizes data sample through 5 features of the data set: minimum,maximum, median, lower quartile, upper quartile. The Box-and-Whisker diagramalso indicates the presence of outliers.

The authors suppose that the tested strength values of bamboo samples shouldbe equal regardless of the position/ height of the cut and that strength value differ-ences of 20–30 N/mm2 are not significant with respect to the aforementionedpotential use. Besides the calculation of statistical features, standardized skewness

Table 5Summary statistics of static tensile test for Tonkin Cane.

Ts_dTC Ts_wTC

Number of samples 5 6Average 210,4 172,65Median 215,27 180,33Variance 672,13 3431,0Standard deviation 25,925 58,575Coeff. of variation 12,322% 33,928%Minimum 182,26 103,06Maximum 246,16 234,62Lower quartile 188,09 114,25Upper quartile 220,22 223,29Interquartile range 32,13 109,04Stnd. skewness 0,29259 �0,16786Stnd. kurtosis �0,46899 �13385

t-test of tensile strength (a = 0,05)Null hypothesis: Ho

Ts_dTC = Ts_dTC_ median

Alternative hypotesis: H1

Ts_dTC– Ts_dTC_ median

t-value �0,39675p-value 0,71180Accept Ho Yes

Comparison of Means95,0% confidence interval for mean 210,4 ± 32,1908

[178,209;242,591]172,647 ± 61,4708[111,176;234,117]

95,0% confidence interval for the difference betweenthe means assuming equal variances

37,7533 ± 64,3204 [�26,5671; 102,074]

t-test to compare means (a = 0,05)Null hypothesis: Ho

Ts_dTC_ mean = Ts_wTC_mean


Ts_dTC_ mean– Ts_wTC_mean

t-value 1,32779p-value 0,216936Accept Ho Yes

Table 6Summary statistics of static tensile test for Ku Zhu.

Ts_dKZ Ts_wKZ

Number of samples 5 5Average 193,58 155,34Median 194,22 150,02Variance 33,726 973,47Standard deviation 58074 31,201Coeff. of variation 30001% 20,085%Minimum 186,96 113,7Maximum 199,77 199,18Lower quartile 188,35 146,79Upper quartile 198,58 167,03Interquartile range 10,23 20,24Stnd. skewness �0,13422 0,16072Stnd. kurtosis �12634 03973

t-test of tensile strength (a = 0,05)Null hypothesis: Ho

Ts_dKZ = Ts_dKZ_ median


Ts_dKZ– Ts_dKZ_ median

t-value -0,16325p-value 0,87823Accept Ho Yes


[186,365;200,787]155,344 ± 38,7407[116,603;194,085]

95,0% confidence interval for the difference between themeans assuming equal variances

38,232 ± 32,7291 [5,50291;70,9611]


Ts_dKZ_ mean = Ts_wKZ_mean


Ts_dKZ_ mean– Ts_wKZ_mean

t-value 2,69373p-value 0,0273382Accept Ho No


0

50

100

150

200

250

300

1 2 3 4 5 6

R m, N

/mm

2

specimen

Tonkin Cane - dry

Tonkin Cane - wet

Ku Zhu - dry

Ku Zhu - wet

Fig. 6. The tensile strength curves for Tonkin Cane and Ku Zhu bamboo in dry and wet condition.

Fig. 7. Fracture mode: a) Dry condition – fracture in narrowest section. (b) Wet condition – interlaminar fracture.

Fig. 8. Box and Whisker Plot of static tensile test for Tonkin Cane.

Fig. 9. Box and Whisker Plot of static tensile test for Ku Zhu.


and standardized kurtosis which give data on strength distribution (asymmetry)around the mean or median value, the aforementioned is proven by tested hypoth-esis (t-test, a = 0.05) that the bamboo samples have strength which equals the med-ian value of the population in dry condition.

3. Results and discussion

3.1. The weight of specimens after 3 weeks, gravimetric methods

Culm specimens were exposed in wet chamber with relativehumidity (60 ± 2)% and temperature (20 ± 2)�C for 21 days andafter that they were weighted on KERN ABS and METTLER TOLEDOdevices. The measured and calculated data are in Table 2 where thesymbols are as follows: weight before – the weight of culm speci-men before wet camber, gram; weight after – the weight of culmspecimen after wet chamber, gram,Dm – weight loss of culm spec-imen i.e. the moisture weight that was accumulated during thetime in the wet camber, gram, w (H2O) – the weight of accumu-

lated moisture (it is not the total weight of moisture in bamboobecause despite the drying the specimen still contains moisture)and the change of moisture content in specimens, in percentage.

The change of moisture content in specimens is between 0,58and 0,73% for Tonkin Cane and 0,39–0,63% for Ku Zhu. It is a verylow percentage and therefore there is no significant change ofweight of bamboo specimens after exposure to the defined wetcondition in wet camber.

3.2. Tensile test

Tables 3 and 4 show the results for both materials in the exam-ination conditions (dry and wet) where the symbols are as follows:a0- cross sectional thickness, mm, b0- cross sectional width, mm,S0- cross sectional area, mm2, Fm- maximum force, kN, Rm- tensilestrength, N/mm2.

According to Tables 3 and 4 and Fig. 6 Tonkin Cane enduredhigher force than Ku Zhu bamboo in the wet and dry condition

Table 7Results of compressive strength test for Tonkin Cane.

No. Sample d1 mm d2 mm S0 mm2 Fm kN rt N/mm2

Dry condition (Cs_dTC)1. 1-1 62,8 53,0 891,303 69,23 77,6732. 1-2 62,7 53,5 839,624 65,15 77,5943. 1-3 62,5 52,8 878,399 62,90 71,6084. 1-4 61,8 52,2 859,542 62,33 72,5155. 1-5 61,7 52,5 825,173 76,61 92,841

Average 62,3 52,8 858,8 67,20 78,400

Wet condition (Cs_wTC)1. 1-A-T 60,30 50,96 816,16 44,57 54,612. 1-A-2T 60,43 51,40 793,12 45,09 56,853. 1-A-3T 61,01 50,62 910,93 70,75 77,674. 1-A-4T 60,62 50,92 849,75 39,09 46,005. 1-A-5T 61,23 50,58 935,23 52,41 56,04

Average ? 60,72 50,90 861,038 50,38 58,23

Table 8Results of compressive strength test for Ku Zhu.

No. Sample d1 mm d2 mm S0 mm2 Fm kN rt N/mm2

Dry condition (Cs_dKZ)1. 18-T-1 40,5 31,5 508,939 43,140 84,7652. 18-T-2 40,0 32,1 447,356 40,880 91,3813. 18-T-3 40,0 31,8 462,412 38,240 82,6974. 18-T-4 39,8 32,0 439,855 37,800 85,9375. 18-T-5 40,0 31,9 457,409 38,950 85,154

Average 40,1 31,9 463,194 39,802 85,987

Wet condition (Cs_wKZ)1. 22-B-T 39,65 27,80 627,76 43,98 70,062. 22-B-2T 39,11 27,08 625,39 40,10 64,123. 22-B-3T 39,55 27,48 635,43 45,93 72,284. 22-B-4T 39,34 28,38 582,93 46,32 79,465. 22-B-5T 39,21 26,78 644,23 42,41 65,83

Average ? 39,37 27,50 623,15 43,75 70,35

0

20

40

60

80

100

1 2 3 4 5

σ t, N

/mm

2

Specimen

Tonkin Cane - dry

Tonkin Cane - wet

Ku Zhu - dry

Ku Zhu - wet

Fig. 10. The compressive strength curves for Tonkin Cane and Ku Zhu bamboo in dry and wet condition.


and 8% higher value of tensile strength than Ku Zhu bamboo in drycondition. Both bamboos show lower values of tensile strengthafter they were exposed to high humidity: Tonkin Cane by 17,9%and Ku Zhu by 19,75%. Typical force-crosshead position curve oftest specimen are shown in Fig. 3b. The difference of fracture modeof the specimens in dry and in wet condition is shown in Fig. 7.

In the wet condition 2/3 of specimens (1-A-3, 1-A-4 and 1-A-6 –Tonkin Cane, and 22-B-1, 22-B-3, 22-B-4 and 22-B-5 – Ku Zhu) lon-gitudinally split before they transversely crack. According to [11]the crack growth along inter-laminar boundaries in the tensile test,it is possible that moisture weakens connections between fibersand causes faster crack growth along inter-laminar boundariesand finally the fracture of specimen.

3.2.1. Tensile test statistics for Tonkin Cane(Figs. 8 and 9)

3.2.2. Tensile test statistics for Ku Zhu

3.3. Compression test

Specimens were tested under compressive load until failure asshown in Fig. 4d. Results of the compressive strength tests areshown in Tables 7 and 8 where symbols are as follows: d1 – exter-nal diameter, mm, d2 – internal diameter, mm, S0 – cross-sectionalarea, mm2, Fm – maximum force, kN and rt – compressive strength,N/mm2.

Fig. 11. (a) Specimen with node. (b) Specimen after testing.

Fig. 12. Box and Whisker Plot of compression test for Tonkin Cane.

Fig. 13. Box and Whisker Plot of compression test for Ku Zhu.


Results of the maximum force and the compressive strengthwith respect to the external and internal diameter were given inTables 7 and 8 and in Fig. 10. Comparing the maximum force andthe compressive strength of Tonkin Cane and Ku Zhu in dry envi-ronmental condition, Tonkin Cane has higher values of maximumforces but, because of distinction in diameter, lower values of com-pressive strength than Ku Zhu. The compression capacity in normalcondition is found to be at its maximum at about 90 kN for TonkinCane and Ku Zhu. This result is 6% higher than in other specimensbecause of presence of node in specimen 1–5 (Fig. 11a) and 18-T-2.

The node in the specimen kept the fibers together and thus delayedthe cracking [12] (Fig. 12 and 13).

According to Tables 7 and 8 and Fig. 11 the compressivestrength of Tonkin Cane and Ku Zhu specimens in dry conditionare 78,4 and 85,98 N/mm2, respectively. However, compared totheir value in dry condition, the average strength in wet conditionhas decreased by 25,7% for Tonkin Cane and by 18% for Ku Zhuamounting to 58,23 and 70,35 N/mm2,, respectively. The failuremode in both conditions was splitting shown in Fig. 4c. with cracksgoing along longitudinal fibers of bamboo under the finite load.Typical force-crosshead position curve of test specimen are shownin Fig. 4b.

3.3.1. Compression test statistics for Tonkin Cane

3.3.2. Compression test statistics for Ku Zhu

3.4. Bending test

The compression stresses in the bending test causes strain per-pendicular to the fibers. This occurs in the material between thefibers which are weak in taking strain. The fibers are still in goodcondition despite the strain on material between them (lignin)but coherence in the cross-section is lost and the bamboo culmdrops very fast. If the load is removed before overloading, the spec-imen will return to its original straight form – a definite advantagefor bike frames.

Bending results for both materials are given in Tables 11 and 12and in Fig. 14. Where the bending strength is given by rS =My/Iand the symbols are as follows: rs – bending strength, N/mm2, M– moment at mid-span [Nmm], y – distance from the surface tothe centre of specimen, mm and I – polar moment of inertia,mm4. In the Tables 10 and 11 they are: l – length of specimen[mm], r0 = y – external radius, mm, ri – internal radius mm, d1 –external diameter, mm and Fm – maximum force, N (Figs. 15 and16).

In bending perpendicular to fibers in bamboo specimens, bothmaterials (Tonkin Cane and Ku Zhu) demonstrated failure by split-ting (Fig. 5c). Typical force-crosshead position curve of test speci-mens may be found in Fig. 5b. Results for both materials havehigher values in dry condition than in wet condition. The bendingstrength for Tonkin Cane is 28,42 N/mm2 and for Ku Zhu is36,85 N/mm2 in dry condition. In wet condition, bending strengthfor both materials is reduced; for Tonkin Cane roughly by ¼ of itsoriginal value to 20,28 N/mm2 and for Ku Zhu by half i.e. to18,54 N/mm2.

Table 9Summary statistics of compression test for Tonkin Cane.

Cs_dTC Cs_wTC

Number of samples 5 5Average 78,446 63,302Median 77,594 57,18Variance 72,619 102,17Standard deviation 85217 10,108Coeff. of variation 10,863% 15,968%Minimum 71,608 54,61Maximum 92,841 77,67Lower quartile 72,515 56,85Upper quartile 77,673 70,2Interquartile range 5158 13,35Stnd. skewness 15079 0,78804Stnd. kurtosis 13644 �0,71496

t-test of compression strength (a = 0,05)Null hypothesis: Ho

Cs_dTC = Cs_dTC_ median


Cs_dTC– Cs_dTC_ median


Comparison of Means95,0% confidence interval for mean 78,4462 ± 10,5811 [67,8651;89,0273] 63,302 ± 12,5507 [50,7513; 75,8527]95,0% confidence interval for the difference between

the means assuming equal variances15,1442 ± 13,6343 [1,50991; 28,7785]


Cs_dTC_ mean = Cs_wTC_mean


Cs_dTC_ mean– Cs_wTC_mean


Table 10Summary statistics of compression test for Ku Zhu.

Cs_dKZ Cs_wKZ

Number of samples 5 5Average 85,987 67,582Median 85,154 66,15Variance 10,527 11,864Standard deviation 32446 34444Coeff. of variation 37733% 50966%Minimum 82,697 64,12Maximum 91,381 72,28Lower quartile 84,765 65,3Upper quartile 85,937 70,06Interquartile range 1172 4,76Stnd. skewness 13246 0,58262Stnd. kurtosis 13243 �0,81471

t-test of compression strength (a = 0,05)Null hypothesis: Ho

Cs_dKZ = Cs_dKZ_ median

Alternative hypothesis: H1

Cs_dKZ– Cs_dKZ_ median



[81,9581; 90,0155]67,582 ± 4,27679[63,3052; 71,8588]

95,0% confidence interval for the difference betweenthe means assuming equal variances

18,4048 ± 4,87993 [13,5249; 23,2847]


Cs_dKZ_ mean = Cs_wKZ_mean


Cs_dKZ_ mean– Cs_wKZ_mean



0

10

20

30

40

50

1 2 3 4 5

σS, N

/mm

2

Specimen

Tonkin Cane - wetKu Zhu - dryKu Zhu - wetTonkin Cane - dry

Fig. 14. The bending strength curves for Tonkin Cane and Ku Zhu bamboo in dry and wet condition.

Fig. 15. Box and Whisker Plot of bending test for Tonkin Cane.

Fig. 16. Box and Whisker Plot of bending test for Ku Zhu.


3.4.1. Bending test statistics for Tonkin Cane

3.4.2. Bending test statistics for Ku ZhuThis paper has presented valuable and significant results. They

indicate that mechanical properties of Tonkin Cane and Ku Zhubamboo in dry (environmental) condition are acceptable andwithin the range of requirements for bicycle frames. The tensilestrengths of tested specimens in dry condition are lower than tita-nium alloy or carbon-fiber composite but they are higher thansome aluminium alloy or bamboo species that were studied insome prior research [5,6,15]. Since water content is a very impor-

tant factor of mechanical properties of organic materials, this studyhas investigated the influence of humidity on compression, tensionand bending strength of bamboo samples after 21 days in wetchamber. The results show that mechanical properties of TonkinCane and Ku Zhu in wet condition are different from those in drycondition. In spite of low percentage of change in moisture content,the values of strengths in wet condition are lower than in dry. Asshown in Figs. 6, 10 and 14 tensile, compressive and bendingstrength appear to decrease with long exposure to wet condition.The tensile strength is decreased by 17,9% for Tonkin Cane andby 19,75% for Ku Zhu in wet condition. The compressive strengthis decreased by 25,7% for Tonkin Cane and by 18% for Ku Zhu ascompared to their values in dry condition. The bending strengthis decreased by 28,4% for Tonkin Cane and by 50% for Ku Zhu inwet condition.

The cause of this change may be attributed to changes ofmicrostructure of bamboo and the influence of humidity onmicrostructure of bamboo should be investigated in some furtherstudies. A common failure mode of culm specimens in bendingand compressive tests is longitudinal splitting failure and it occursin both tests, in dry and wet condition. Wet samples in tensile testcreate faster longitudinal splitting, before they crack transversely.Because of this there might be changes in the microstructure, withpossible weakening of connection between fibers.

For a more direct confirmation, mechanical properties of twostructural bamboo for bamboo scaffoldings were tested in dryand wet condition in work [18]. The specimens in wet tests wereimmersed in water over different time periods. This study confirmsthat the influence of humidity is very important and that bambooexposure to high humidity leads to a decrease in the followingmechanical properties: bending strength is reduced by 30% andcompression strength by 50% of their original value in dry condi-tion. Some other study [22] shows that bamboo exposed for ashorter time (about 48 h) under high humidity (60%) has moreductility under torsion. Having looked into scientific articles ofrecent date no relevant studies dealing with the influence of mois-ture on the microstructure and mechanical properties of bamboohave been found.

Therefore, the results of this study and some other studies[18,22] show that there is a need for more detailed research intobehaviour of bamboo exposed to high humidity over different timeperiods.

The results presented in this work suggest that the durability ofbamboo depends on better protection treatment methods (impreg-nation) of bamboo. It should be taken into consideration in thedesign of bicycle frames using bamboo and the storage of builtbicycles should be taken into account during wet days.

Table 11Results of bending strength test for Tonkin Cane.

No. Sample l mm r0 mm ri mm Fm N d1 mm M Nmm I mm4 rs N/mm2

Dry condition (Bs_dTC)1. 13-1 197 11,60 8,25 990 23,20 18810 10582,37 20,622. 13-2 199 11,05 8,60 970 22,10 18430 7413,33 27,473. 13-3 202 11,05 8,85 1160 22,10 22040 6891,56 35,344. 13-4 200 10,97 9,02 1016 21,33 19306 6175,13 34,305. 13-5 199 11,09 8,16 972 22,31 18470 8397,81 24,39

Average ? 199 11,15 8,58 1021,64 22,21 19411 7892,04 28,42

Wet condition (Bs_wTC)1. 9-A-1 198 13,25 9,35 1640 26,50 31160 18205,14 22,682. 9-A-2 196 13,70 8,86 1610 27,40 30590 22827,87 18,363. 9-B-1 200 13,22 9,48 1500 26,43 28500 17645,78 21,354. 9-B-2 198 13,51 9,40 1653 26,89 31407 20032,44 21,185 9-B-3 197 13,40 8,59 1475 26,84 28025 21046,30 17,84

Average ? 198 13,12 9,14 1576 26,81 29936 19951,51 20,28

Table 12Results of bending strength test for Ku Zhu.

No. Sample l mm r0 mm ri mm Fm N d1 mm M Nmm I mm4 rs N/mm2

Dry condition (Bs_dKZ)1. 25-1 201 12,06 7,90 2440 24,10 47310 13555,06 42,092. 25-2 198 11,60 7,93 1870 23,20 35530 11114,86 37,083. 25-3 201 11,60 7,85 1510 23,20 28690 11238,31 29,614. 25-4 202 11,59 7,95 1541 23,34 29279 11034,44 30,755. 25-5 199 11,42 7,89 2127 23,62 40413 10314,72 44,74

Average ? 200 11,65 7,90 1897 23,49 36244 11451,48 36,85

Wet condition (Bs_wKZ)1. 24-A-1 231 13,25 9,35 2490 22,91 18934 18205,10 13,782. 24-A-2 199 13,70 8,86 1880 23,15 15580 22827,80 9,353. 24-B-1 196 13,22 9,48 1500 24,19 22420 17645,70 16,804. 24-B-2 237 13,28 8,60 2029 23,07 38551 20131,49 25,435. 24-B-3 220 13,31 9,62 1936 22,26 36784 17922,64 27,32

Average ? 217 13,35 9,18 1967 23,12 26454 19346,55 18,536

Table 13Summary statistics of bending test for Tonkin Cane.

Bs_dTC Bs_wTC

Number of samples 5 5Average 28,424 20,282Median 27,47 21,18Variance 40,111 43387Standard deviation 63333 2083Coeff. of variation 22,282% 10,27%Minimum 20,62 17,84Maximum 35,34 22,68Lower quartile 24,39 18,36Upper quartile 34,3 21,35Interquartile range 9,91 2,99Stnd. skewness �0,012149 �0,24961Stnd. kurtosis �10079 �10987

t-test of bending strength (a = 0,05)Null hypothesis: Ho

Bs_dTC = Bs_dTC_ median


Bs_dTC– Bs_dTC_ median



[20,5601; 36,2879]20,282 ± 2,58634[17,6957; 22,8683

95,0% confidence interval for the difference between the means assuming equal variances : 8142 ± 68756 [12664; 15,0176]


Bs_dTC_ mean = Bs_wTC_mean


Bs_dTC_ mean– Bs_wTC_mean



Table 14Summary statistics of bending test for Ku Zhu.

Bs_dKZ Bs_wKZ

Number of samples 5 5Average 36,854 18,536Median 37,08 16,8Variance 44,848 58,675Standard deviation 66968 7,66Coeff. of variation 18,171% 41,325%Minimum 29,61 9,35Maximum 44,74 27,32Lower quartile 30,75 13,78Upper quartile 42,09 25,43Interquartile range 11,34 11,65Stnd. skewness 0,033462 0,099409Stnd. kurtosis �11663 �10242

t-test of bending strength (a = 0,05)Null hypothesis: Ho

Bs_dKZ = Bs_dKZ_ median


Bs_dKZ– Bs_dKZ_ median

t-value �0,04874p-value 0,96345Accept Ho Yes


[28,5388; 45,1692]18,536 ± 9,51116[9,02484; 28,0472]

95,0% confidence interval for the difference between the means assuming equal variances 18,318 ± 10,4929 [7,82513; 28,8109]


Bs_dKZ_ mean = Bs_wKZ_mean

Alternative hypothesis: H1 Bs_dKZ_ mean– Bs_wKZ_mean



4. Conclusion

The study on two bamboo species that are suitable for bikeframes was carried out to examine the influence of humidity onmechanical properties of bamboo. The experimental results indi-cate that moisture is a very important factor in mechanical proper-ties of bamboo. The results indicate that after exposing specimensto the defined wet condition the percentage of moisture change inspecimens is less than 0,73% in both materials and therefore thereis no significant change of weight of bamboo specimens. In spite oflow changes in the weight of samples, the overall results show thatmechanical properties are lower than in dry condition. The resultsshow that the tensile, compressive and bending strength of bam-boo have significantly decreased after the bamboo samples werekept in an environment with humidity level of 60% for 3 weeks.

At the level of significance a = 0.05, it has been proven thatthere is no significant difference in the amount of tensile strengthin dry and wet condition of Tonkin Cane samples. P-value in thiscase is p = 0.216936, greater than 0.05 and thus accepting thehypothesis H0 of mean equality tensile strength between dry andwet samples. Means of Ku Zhu patterns show there is a statisticallysignificant difference between the mean of tensile strength of sam-ples in dry and wet state (p = 0.0273382) and therefore the hypoth-esis H0 is rejected.

It is also proven at the level of statistical significance that thereis a significant difference between the mean of compressivestrength in dry and wet condition of Tonkin Cane (p = 0.0335746)and Ku Zhu (p = 0.0000238) samples.

A statistical comparison of mean bending strength of TonkinCane (p = 0.025816) and Ku Zhu (p = 0.003810) samples also showsa significant difference in dry and wet condition.

In this study, the sampling was intentionally done by randomchoice of the height i.e. position of cuts of the samples. The authorssupposed that the tested strengths of bamboo samples should beequal regardless of the position of the cut. According to the

authors, differences in strength values in dry condition, whichobviously exist, are not important (e.g. 20–30 N/mm2) withrespect to the aforementioned potential use. Considering theamounts of statistic features, standardized skewness and standard-ized kurtosis (within the range �2 to +2) it is evident that strengthdistribution in dry condition is within normal strength distributionlimits suggesting that strength values are accumulating around themean or median values i.e. that they are more or less uniformregardless of the position (height) of sampling i.e. cutting. Theabove is proven by tested hypothesis (t-test, a = 0.05) that strengthvalues of individual samples equal the median value of samplesfrom the population (Tables 5, 6, 9, 10, 13 and 14).

The results of the study indicate that the time of exposure tohumidity may be useful in construction of bicycle frames in orderto take advantage of the differences in strength between dry andwet conditions.

Further research is necessary to fully explain the changes inmechanical properties of bamboo after exposure in wet conditionfor a long period and to determine the connection on the micro-level. Additional research may lead to dynamic tests of bamboofor bicycles. Future research is planned to test dynamic strengthin dry and wet conditions and development of mathematical mod-els for predicting the amount of dynamic strength in wet condition,depending on the time of exposure to humidity.

References

[1] B. Sharma, A. Gatoo, M.H. Ramage, Effect of processing methods on themechanical properties of engineered bamboo, Constr. Build. Mater. 83 (2015)95–101.

[2] Y. Xiao, Q. Zhou, B. Shan, Design and construction of modern bamboo bridges, J.Bridge Eng. 15 (5) (2010) 533–541.

[3] C. Esteve-Sendra, R. Moreno-Cuesta, A. Portales-Mananos, T. Magal-Royo, Bamboo, from traditional crafts to contemporary design and architecture, ProcediaSoc. Behav. Sci. 51 (2012) 777–781.

[4] Scurlock J.M.O, D.C. Dayton, B. Hames, Bamboo: an overlooked biomass re-source?, in: Biomass Bioenergy 19 (2000) 229–244.

http://refhub.elsevier.com/S0950-0618(17)31087-5/h0005











[5] H.P.S. Abdul Khalil, I.U.H. Bhat, M. Jawaid, A. Zaidon, D. Hermawan, Y.S. Hadi,Bamboo fibre reinforced biocomposites: a review, Mater. Des. 42 (2012) 353–368.

[6] Z. Aiping, H. Dongsheng, L. Haitao, Y. Su, Hybrid approach to determine themechanical parameters of fibers and matrixes, Constr. Build. Mater. 35 (2012)191–196.

[7] R. Van der Plas, S. Baird, Bicycle Technology: Understanding the ModernBicycle and its Components (Cycling Resources), second ed., Cycle publishing,San Fran-cisco, 2010.

[8] C. Haine, The Urban Biking Handbook: The DIY Guide to Building, Rebuilding,Tinkering with, and Repairing Your Bicycle for City Living, first ed., QuarryBooks, Massachusett, 2011.

[9] W. Liese, M. Köhl, Bamboo – The Plant and its Uses, Springer InternationalPublishing, Switzerland, 2015.

[10] W. Schott, Bamboo in the Laboratory, http://www.powerfibers.com/BAMBOO_IN_THE_LABORATORY. pdf, 2006 (accessed 30.11.15).

[11] T. Tan, N. Rahbar, S.M. Allameh, S. Kwofie, D. Dissmore, K. Ghavami, W.O.Soboyejo, Mechanical properties of functionally graded hierarchical bamboostruc-tures, in: Acta Biomater. 7 (2011) 3796–3803.

[12] J.A. Janssen, Designing and Building with Bamboo, Technical report no. 20 –Technical University of Eindhoven Netherlands, International Network forBamboo and Rattan 2000.

[13] M. Ahmad, Mechanical Properties of Calcutta Bamboo (Ph.D. thesis), VirginiaPolytechnic Institute and State University, Blacksburg USA, 2000.

[14] M.R. Gerhardt, Microstructure and Mechanical Properties of Bamboo inCompres-sion (Bachelor Thesis), Massachusetts USA, Massachusetts Instituteof Technology, 2012.

[15] T. Gutu, A study on the mechanical strength properties of bamboo to enhanceits diversification on its utilization, Int. J. Innov. Technol. Explor. Eng. 2 (2013)314–319.

[16] C. Okhio, J. Waning, Y.T. Mekonnen, An Experimental Investigation of the Eff-ects of Moisture Content on the Mechanical Properties of Bamboo and Cane,Cyber Journals: Multidiscip. J. Sci. Technol., J. Sel. N.A. Bioeng. (2011) 7–14.

[17] B. Sharma, A. Gatóo, M. Bock, M. Ramage, Engineered bamboo for structuralapplications, Constr. Build. Mater. 81 (2015) 66–73.

[18] K.F. Chung, W.K. Yu, Mechanical properties of structural bamboo for bambooScaffoldings, Eng. Struct. 24 (2002) 429–442.

[19] T.Y. Lo, H.Z. Cui, H.C. Leung, The effect of fiber density on strength capacity ofbamboo, Mater. Lett. 58 (2004) 2595–2598.

[20] S. Amada, S. Unttao, Fracture properties of bamboo, Compos. B 32 (2001) 451–459.

[21] H. Liu, Z. Jiang, B. Fei, C. Hse, Z. Sun, Tensile behaviour and fracture mechanismof moso bamboo (Phyllostachys pubescens), Holzforsch 69 (2015) 47–52.

[22] S. Askarinejad, P. Kotowski, F. Shalchy, N. Rahbar, Effects of humidity on shearbehavior of bamboo, Theor. Appl. Mech. Lett. 5 (2015) 236–243.

[23] S.C. Lakkad, J.M. Patel, Mechanical properties of bamboo, a natural composite,Fibre Sci Technol. 14 (1981) 319–322.


















http://www.powerfibers.com/BAMBOO_IN_THE_LABORATORY

http://www.powerfibers.com/BAMBOO_IN_THE_LABORATORY




























the influence of humidity on mechanical properties of...

Documents