Materials and Design 105 (2016) 152–159

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

A novel approach to improve mechanical properties of parts fabricated by

fused deposition modeling
Jianlei Wang a,c, Hongmei Xie a, Zixiang Weng a,c, T. Senthil a, Lixin Wu a,b,⁎
a
Key Laboratory of design and assembly of functional nanostructures, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou 350002, China
b
Fujian Provincial Key Laboratory of Nanomaterials, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou 350002, China
c
University of Chinese Academy of Sciences, Beijing 100049, China

a r t i c l e i n f o a b s t r a c t

Article history: The mechanical properties of fabricated parts by fused deposition modeling (FDM) are impaired by significant
Received 6 April 2016 voids between deposition lines. In this study, a novel approach that adding thermally expandable microspheres
Received in revised form 18 May 2016 into matrix and combining FDM process with thermal treatment was proposed to tackle the problem. The influ-
Accepted 19 May 2016
ences of microsphere content, heating temperature and heating time were investigated. The results show that
Available online 20 May 2016
appropriate conditions of thermal treatment exert a positive influence on mechanical properties. With the addi-
Keywords:
tion of 2 wt.% microspheres, tensile and compressive strength of samples heated at 140 °C for 120 s increase 25.4%
Fused deposition modeling and 52.2%, compared to untreated samples, respectively. A new process is developed to prepare foam materials
Thermally expandable microsphere based on FDM, which will have the potential applications in shoes industry.
Thermal treatment © 2016 Elsevier Ltd. All rights reserved.
Mechanical properties
Foam materials

1. Introduction extruded material cools quickly from melting temperature to chamber

temperature, resulting in development of inner stresses responsible
Additive manufacturing (AM) has captured the world's horizon for weak bond between two deposition lines, which leads to inter-
since Charles Hull firstly invented the new manufacturing technique and intra-layer deformation in the form of cracking, delamination or
called stereo lithography (SLA) in 1984. As the computer processing even part fabrication failure [11].
technologies make breakthroughs, AM is developing at an amazing In recent years, researchers mainly worked towards two directions
pace in the speed and accuracy of printing, the type and property of ma- and tried to improve the mechanical properties of fabricated parts.
terial and manufacturing cost [1]. Fused deposition modeling (FDM) Some researchers investigated the nature of FDM and the effects of pro-
invented and developed by Stratasys Inc. in the early 1990s is the cess parameters and algorithm. Sun. et al. [12] experimentally demon-
most widely used technique among all AM technologies, showing high strated that bond quality between adjacent filaments depended on
potentials for fabricating plastic parts with the capacity to compete envelope temperature and variations in the convective conditions with-
with conventional processing techniques [2,3]. The range of applica- in the building part. Qiu and Langrana [13] studied the toolpath effects
tions of FDM is extensive, ranging from medical treatment [4,5], in FDM and proposed an algorithm to match the toolpath with the ex-
mould design [6] to automotive [7] and aeronautics [8]. trusion speed to eliminate voids and to correct overfill and underfill de-
In the FDM process, the thermoplastic filament as feedstock is fed fects. Zhang and Chou [14] conducted a parametric study of part
into heating chamber by a stepping motor and extruded through the distortions in FDM using 3D FEA. The simulation results showed that
nozzle in a prescribed manner upon its melting temperature, then solid- the scan speed was the most significant factor to part distortions follow-
ified and deposited on the platform on a layer by layer basis [9,10]. Al- ed by the layer thickness. Furthermore, they reported that residual
though the deposition lines can be integrated into adjacent deposition stresses increased with layer thickness and deposition line's width to a
lines a bit by its own gravity and the force of stepping motor, there less extent. Antreas Kantaros and Dimitris Karalekas [15] investigated
exist significant voids between deposition lines, which impairs the me- the effects of layer thickness and deposition orientation on the residual
chanical properties of fabricated parts to a great degree. Also the strains at the end of the fabrication process recorded by an optical sen-
sor with a short fiber Bragg grating (FBG) embedded at the midplane of
FDM built samples.
⁎ Corresponding author at: Key Laboratory of design and assembly of functional
nanostructures, Fujian Institute of Research on the Structure of Matter, Chinese Academy
Meanwhile, other researchers studied the problem from the aspect
of Sciences, Fuzhou 350002, China. of materials, adding fibers or particles into thermoplastic matrix to im-
E-mail address: lxwu@fjirsm.ac.cn (L. Wu). prove mechanical strength and modulus. Gray IV et al. [16] added

http://dx.doi.org/10.1016/j.matdes.2016.05.078
0264-1275/© 2016 Elsevier Ltd. All rights reserved.
 J. Wang et al. / Materials and Design 105 (2016) 152–159 153

Fig. 1. Types of specimens printed by FDM: (a) tensile; (b) compressive; (c) section topography of black rectangle plane in (a) and (b).

Fig. 2. The process flow chart of fabrication of specimens.

154 J. Wang et al. / Materials and Design 105 (2016) 152–159

thermotropic liquid crystal-line polymer fibrils into polypropylene in

order to prepare a composite feedstock for FDM. A capillary rheometer
was used to simulate the FDM process, and subsequently the tensile
property of extruded strands was found to improve with draw ratio.
Zhong et al. [17] studied FDM process of short glass fiber reinforced
ABS. Additions of plasticizer and compatibilizer improved feedstock
processibility. Shofner et al. [18] investigated the effects of vapor-
grown carbon fibers added into ABS as FDM feedstock on mechanical
properties. An average of 39% increase in tensile strength was observed
at 10 wt.% loading of nanofiber. Halil L. Tekinalp et al. [8] reinforced ABS
by adding short carbon fiber to investigate its processibility, microstruc-
ture and mechanical performances. The tensile strength and modulus of
3D-printed samples increased 115% and 700%, respectively and the car-
bon fiber orientation of fabricated samples was up to 91.5% in the print-
ing direction.
In this study, we came up with a novel approach that making use of
the expandability of thermally expandable microspheres to reduce the
voids between deposition lines so as to improve the mechanical proper-
ties of fabricated parts. Also a new process to prepare foam materials
based on FDM was developed.

2. Experimental work

2.1. Materials and processing

Thermally expandable microspheres are often utilized in compres-

sion, injection molding and extrusion processes to foam different
types of polymers. Microspheres consist of a thermoplastic polymer
shell and a liquid hydrocarbon encapsulated inside with low boiling
point, such as octane or pentane [19]. When microspheres are heated,
the internal pressure increases dramatically and shell goes soft. As a re-
sult, they expand to 50–100 times of initial volumes. In this study, ther-
mally expandable microspheres named Expancel 930DU120 from
AkzoNobel were used, whose initial foaming temperature was about
120 °C. To satisfy its processibility and ensure that microspheres do Fig. 3. Effects of microsphere content on tensile and compressive strength.
not expand during extrusion and FDM process, the melting point of ma-
trix must be below 120 °C. In that case, we developed Polywax based on
PE wax fitting for FDM process [20]. The melting temperature of treatment in a press. The silk net with low thermal conductivity
Polywax was about 95 °C tested by DSC and the melt flow index was was mounted between press's plates with the distance of 15 mm so
19 g/10 min under the condition of 0.325 kg, 125 °C, according to as to evenly heat the samples without directly touching the press.
ISO1133. Polywax has low linear heat shrinkage rate, so the fabricated After the early exploration, the range of heating temperature and
part has no warpage and accurate dimensional stability. heating time was determined in case of deformation in shape within
The filament fabrication contained two procedures. Firstly, micro- the scope that naked eyes can distinguish. The temperature of 130–
spheres and Polywax were mixed for 10 min and then compounded in 180 °C with the interval of 10 °C and the heating time of 40–120 s
HAAKE twin-screw extruder to prepare masterbatch. The content of mi- with the interval of 20 s were employed as the conditions of thermal
crospheres was 20 wt.%. Secondly, the masterbatch and Polywax were treatment. The whole process is illustrated in Fig. 2.
mixed and then extruded in a single-screw extruder to prepare filament
with 2.5 mm in diameter. During the process, the barrel temperature
ranged between 75 and 100 °C. Mixtures of 2, 5, 8 and 11 wt.% micro-
spheres were prepared.
Tensile, compressive and DMA specimens were prepared by feed-
ing the filament into a commercial desktop FDM unit provided by
HY3D Corporation. The FDM unit was modified with a PU roller
mounted on the stepper motor to adapt to Polywax. During printing,
nozzle temperature was maintained at 100 °C, while platform tem-
perature was near room temperature. The layer height was set at
0.2 mm and the infill density was set at 100%. The nozzle's diameter
of the FDM unit was 0.5 mm, ensuring that no clogging happened
during printing. The printing speed for the first layer and other layers
were 30 and 50 mm/s, respectively. Because the matrix was flexible
and ductile, compressive test was employed to assess interlaminar
property instead of short beam shear test [21]. ISO 527 5A dog-
bone dimensions and ISO 604 pipe cylinder dimensions were follow-
ed, respectively. Single Cantilever approach was adopted for DMA
test. The shape and cross-section topography of specimens are illus-
trated in Fig. 1. The printed samples were followed by thermal Fig. 4. The effect of microsphere content on storage modulus.
 J. Wang et al. / Materials and Design 105 (2016) 152–159 155

Fig. 5. SEM of fracture interface with (a) 0, (b) 2, (c) 5, (d) 8 and (e) 11 wt.% microspheres.

2.2. Testing and analysis property and to demonstrate that appropriate conditions of thermal
treatment exerted a positive influence on mechanical properties.
ISO 527 standard and ISO 604 standard were followed for tensile test
and compressive test, respectively. Specimens were conducted in a uni- 3.1. Void formation
versal testing machine (AGS-X Shimadzu Co., Kyoto, Japan) with a
1000 N transducer capacity. The cross-head speed for tensile test and To understand the mechanism of void formation, a close observation
compression test were 100 and 5 mm/min, respectively. at the FDM process is necessary. Although the nozzle is circular, the de-
Dynamic mechanical analysis was conducted on DMA Q800 (TA Co., position line is pressed down to a 0.2 mm thickness and it becomes el-
New Castle, USA). The specimens were tested between 25 and 55 °C at a liptical. Because the deposition line is still soft when being deposited,
heating rate of 5 °C/min , with storage modulus G′ recorded with respect the bottom flattens under press, while the top cools to form round
to temperature. edge before another layer is deposited on top of it. It illustrates the rea-
SEM micrographs of fracture surfaces were taken with Hitachi son why the triangular gaps aligned in the direction of printing are only
SU8010 FEG-SEM with an acceleration voltage of 5 kV after specimens downwards, as shown in Fig. 1(c).
were fractured in liquid nitrogen.
3.2. Effects of microsphere content

3. Results and discussion The effect of the content of thermally expandable microspheres was
investigated comparing printed samples with microsphere content of 0,
The purpose of this study was to assess the effects of thermally ex- 2, 5, 8 and 11 wt.%. Samples went through thermal treatment at 140 °C
pandable microspheres added into matrix on tensile and interlaminar for 60 s or 120 s.
156 J. Wang et al. / Materials and Design 105 (2016) 152–159

printing, resulting in smaller voids and thus better adhesion between de-
position lines [8], as observed in Fig. 5. However, when adding more than
8 wt.% microspheres, the dispersion of microspheres becomes difficult
and the agglomeration is significant, which leads to stress concentration
inside the samples [27,28]. Meanwhile, inter-particle voids accumulate.
As a result, compressive strength and storage modulus decrease. The ef-
fects of thermal treatment will be expounded later.

3.3. Effects of heating temperature

In order to study the effect of heating temperature, the printed sam-

ples with 0 and 2 wt.% microspheres were compared. The heating time
for all samples was 60 s.
As shown in Fig. 6, the tensile strength of neat samples increases as
the heating temperature rises, but the tensile strength of samples with
2 wt.% microspheres presents a different trend with a peak at 140 °C, de-
creasing at a relatively high temperature. The trend for compressive
strength is identical to tensile strength, but the curve for samples with
2 wt.% microspheres is entirely above the one for neat samples. It can
be attributed to that heated microspheres expand to push the matrix
to fill the voids, as illustrated in Fig. 7. Furthermore, according to crack
propagation mechanism, cracks always propagate along with the
weakest point when loaded, perpendicular to the loading direction.
The interfaces between two layers can be considered as the crack
paths for untreated samples. However, after being heated, the interface
becomes blurred and cracks propagate along with the expanded micro-
spheres, which lengthens the crack paths and thus consumes more en-
ergy. As a result, mechanical properties improve. Similar results and
explanations were also reported by references [29–31]. When heating
temperature rises to 180 °C, microspheres expand at a dramatic speed
in the marginal areas of samples, because heat cannot transfer into inte-
rior area to melt matrix within a limited period of time, which leads to
stress concentration in the marginal areas, as observed in Fig. 8. In con-
sequence, mechanical properties decrease a lot.
Fig. 9 reveals the variation of storage modulus for varying heating
Fig. 6. Effects of heating temperature on tensile and compressive strength.
temperature. Similar to the explanation for the trend in Fig. 4, reinforce-
ment effect of microspheres is evident. However, polymer chains pos-
sess large free volume at high temperature, resulting in the decrease
As shown in Fig. 3, the influence of microsphere content is significant. in storage modulus [32].
Increasing microsphere content leads to better tensile and compressive
strength, especially the latter ascending from 1.8 MPa to 5.4 MPa, refer-
ring to untreated samples. The same trend is exhibited in all circum- 3.4. Effects of heating time
stances. The result of DMA presents that storage modulus increases
when introducing more microspheres, as exhibited in Fig. 4. This is main- The effect of heating time was evaluated comparing neat samples
ly due to 1) microspheres without expansion have higher strength and and samples with 2 wt.% microspheres, with heating temperature main-
modulus than matrix, working similarly as the reinforced materials, taining at 140 °C. As observed in Fig. 10a, the extension of heating time
such as CaCO3 [22] or OMMT [23]; 2) microspheres restrict the motion is of positive effect. Tensile strength improves when heating time ex-
of molecular chain and improve the overall stiffness, which makes sam- tends, with presenting larger amplification when introducing 2 wt.% mi-
ples endure larger compressive force without deformation and results crospheres, compared to the neat samples. The same trend is exhibited
in better mechanical properties [24,25]; 3) microspheres regarded as nu- in the variation of compressive strength, as shown in Fig. 10b. It can be
cleating agent accelerate crystallization process and increase the degree illustrated that the majority of microspheres expand at an appropriate
of crystallinity of the PE matrix [26]. Moreover, porosity regarded as a cru- size to give polymer matrix impetus to infill the voids after taking in
cial factor for mechanical properties decreases as microsphere content in- enough heat, which further elongates the crack path to endure more
creases, which can be attributed to the reduction in die swell during loading before failure. It has been supported by SEM observation in
Fig. 11. But when adding more than 8 wt.% microspheres, the positive
effect of heating time is less significant (Fig. 3). As explained earlier, it
can be attributed to that after being heated, expanded microspheres fur-
ther agglomerate and stress concentration intensifies as more micro-
spheres expand. Meanwhile, cracks are inclined to grow along with
expanded microspheres [33].
Fig. 12 reveals the effect of heating time on the storage modulus. It is
worth mentioning that the storage modulus of samples with 2% micro-
spheres is nearly two times higher when heating time doubles from 60
Fig. 7. Schematic of thermally expandable microspheres ( purple balls) in matrix (gray to 120 s, which enables to manifest the positive effect of appropriate
area) and crack paths (red lines). conditions of thermal treatment on the stiffness of samples.
 J. Wang et al. / Materials and Design 105 (2016) 152–159 157

Fig. 8. SEM of fracture interface of samples with (a) 0 and (b) 2 wt.% after thermal treatment of 180 °C, 60 s.

3.5. Effects on porosity The porosity of neat samples without thermal treatment is 17.2%,
closing to the analysis result of SEM observation by software in Fig.
There exist three categories of voids in samples, including the vol- 5(a). The effects of varying conditions on porosity, including micro-
ume of expanded microspheres, the gaps between microspheres and sphere content and heating temperature, are shown in Fig. 13. It can
matrix and the physical voids at layer/layer interfaces. The second be observed that the porosity decreases significantly with increasing
type is mainly generated during the fabrication of feedstock filament content of microspheres, which supports the early discussion. The po-
in the extrusion process. While the last type is mainly between adjacent rosity of neat samples decreases slightly as heating temperature rises.
deposition lines, resulting from the FDM process. Therefore, the calcula- However, porosity presents a different trend with adding 2 wt.% micro-
tion of overall porosity should take three types of voids into consider- spheres, which can be attributed to that the augment in volume of ex-
ation and the overall porosity P is expressed by the following equation: panded microspheres counteracts the decrease in voids between
deposition lines, maintaining overall porosity at a certain level.
Vt −Va
P¼ ð1Þ
Vt

where Vt is the volume of fabricated sample and Va is the actual volume

of materials. Although the volume of designed solid object can be direct-
ly calculated by 3D modeling software, the shrinkage of material and
printing accuracy error of FDM machine may affect the result. To mea-
sure Vt accurately, rectangular samples with 2 mm in shell thickness
were prepared to ensure no open pore and Archimedes approach was
adopted. Since there is no reaction between the matrix and micro-
spheres, Va can be calculated as follows:

m  Polywax wt:% m  Microsphere wt:%

Va ¼ þ ð2Þ
ρPolywax ρMicrosphere

where m is the actual mass of the sample; ρPolywax is the density of

Polywax (0.951 g/cm3); ρMicrosphere is the density of microsphere
(0.0065 g/cm3), which can be negligible.

Fig. 9. The effect of heating temperature on storage modulus. Fig. 10. Effects of heating time on tensile and compressive strength.
158 J. Wang et al. / Materials and Design 105 (2016) 152–159

Fig. 11. SEM of fracture interface of samples with (a) 0 and (b) 2 wt.% after thermal treatment of 140 °C, 120 s.

4. Applications of the material interlaminar property. With adding 2 wt.% microspheres, tensile and
compressive strength of samples heated at 140 °C for 120 s increases
Generally, traditional PE foam materials are prepared by following 25.4% and 52.2%, compared to untreated samples, respectively. Current-
methods, including extrusion foaming, tank impregnation, molding ly the materials wildly used as feedstock in FDM are ABS or PLA whose
foaming and injection foaming [34]. It is innovative to take advantage melting temperature is closing to 190 °C. If the thermally expandable
of 3D printing and combine FDM with traditional process to make microsphere can endure higher temperature without expansion,
foam materials with specific or complicated shapes, which has potential whose thermoplastic shell is PI, PPS or PEEK, it visualizes that the me-
applications in shoes industry. Take for example steric printing for the chanical properties of parts fabricated by FDM can be improved in this
design stage in sneaker industry. The designed patterns on the upper way. This study provides the solution and explores the effects of relative
materials are painted layer by layer by workers, which consume time factors, which plays an instructive role. Future work will be focused on
and labor cost. It will make up the defects that incorporating FDM pro- the study of controlling the cellular structure and morphology of the
cess into traditional manufacture, maintaining stereoscopic effect and
hand feeling. The process of steric printing based on FDM is illustrated
in Fig. 14. The fabricated part was of 0.6 mm in height and contained
8 wt.% microspheres. Then it was of about 2.4 mm in height after
being heated at 200 °C for 120 s. The findings of this study facilitate bet-
ter applications of the material. It helps to understand the mechanism of
foaming with microspheres and the effects of thermal treatment so as to
improve the mechanical properties as well as adhesion between mate-
rials and cloth.

5. Conclusions

In this study, a novel approach that adding thermally expandable

microspheres into the matrix and combining FDM process and thermal
treatment was suggested to reduce the voids between deposition lines
so as to improve the mechanical properties. The effects of microsphere
content, heating temperature and heating time on mechanical proper-
ties were investigated. The results indicate that appropriate conditions
of thermal treatment exert a positive influence on tensile and

Fig. 12. The effect of heating time on storage modulus. Fig. 13. Effects of (a) microsphere content and (b) heating temperature on porosity.
 J. Wang et al. / Materials and Design 105 (2016) 152–159 159

Fig. 14. The process of steric printing based on FDM.

foam materials fabricated by FDM and the exploration of process condi- [16] R.W. Gray IV, D.G. Baird, J.H. Bohn, Thermoplastic composites reinforced with long
fiber thermotropic liquid crystalline polymers for fused deposition modeling,
tions to control foaming behavior better. Polym. Compos. 19 (4) (1998) 383–394.
[17] W. Zhong, F. Li, Z. Zhang, L. Song, Z. Li, Short fiber reinforced composites for fused
Acknowledgments deposition modeling, Mater. Sci. Eng. A 301 (2) (2001) 125–130.
[18] M.L. Shofner, K. Lozano, F.J. Rodrı'guez-Macı'as, E.V. Barrera, Nanofiber-reinforced
polymers prepared by fused deposition modeling, J. Appl. Polym. Sci. 89 (11)
This research was financially supported by the National Natural Sci- (2003) 3081–3090.
ence Foundation of China (grant nos.: U1205114 and 51403212) and [19] M. Fujino, et al., Mathematical models and numerical simulations of a thermally ex-
pandable microballoon for plastic foaming, Chem. Eng. Sci. 104 (2013) 220–227.
Science Foundation of Fujian province (grant nos.: 2014J01217 and [20] Jianlei Wang, Lixin Wu, Dongxian Zhuo, Zixiang Weng, Yu Zhou, 2015, CN Pat.
2015H0047). 2015100960566.
[21] Y. Shindo, R. Wang, Theoretical and experimental evaluation of double-notch shear
strength of G-10CR glass-ClothEpoxy laminates at cryogenic temperatures, J. Eng.
References
Mater. Technol. 121 (6) (1999) 367–373.
[22] A. Elloumi, S. Pimbert, C. Bradai, Phase structure and mechanical properties of PP/
[1] D. Raviv, et al., Active printed materials for complex self-evolving deformations, Sci.
EPR/CaCO3nanocomposites: effect of particle's size and treatment, Polym. Eng. Sci.
Rep. 4 (2014) 7422.
55 (12) (2015) 2859–2868.
[2] C. Kousiatza, D. Karalekas, In-situ monitoring of strain and temperature distributions
[23] Z. Weng, et al., Mechanical and thermal properties of ABS/montmorillonite nano-
during fused deposition modeling process, Mater. Des. 97 (2016) 400–406.
composites for fused deposition modeling 3D printing, Mater. Des. 102 (2016)
[3] A. Le Duigou, et al., 3D printing of wood fibre biocomposites: from mechanical to ac-
276–283.
tuation functionality, Mater. Des. 96 (2016) 106–114.
[24] S.-K. Lim, et al., Preparation and interaction characteristics of exfoliated ABS/
[4] P. Gu, L. Li, Fabrication of biomedical prototypes with locally controlled properties
organoclay nanocomposite, Polym. Eng. Sci. 50 (3) (2010) 504–512.
using FDM, CIRP Ann. Manuf. Technol. 51 (1) (2002) 181–184.
[25] M.I. Triantou, P.A. Tarantili, Studies on morphology and thermomechanical perfor-
[5] M. Centola, et al., Combining electrospinning and fused deposition modeling for the
mance of ABS/PC/organoclay hybrids, Polym. Compos. 35 (7) (2014) 1395–1407.
fabrication of a hybrid vascular graft, Biofabrication 2 (1) (2010) 014102.
[26] H.A. Khonakdar, Dynamic mechanical analysis and thermal properties of LLDPE/
[6] S.H. Masood, W.Q. Song, Development of new metal/polymer materials for rapid
EVA/modified silica nanocomposites, Compos. Part B Eng. 76 (2015) 343–353.
tooling using fused deposition modelling, Mater. Des. 25 (7) (2004) 587–594.
[27] A.L. Yesgat, R. Kitey, Effect of filler geometry on fracture mechanisms in glass particle
[7] F. Ning, et al., Additive manufacturing of carbon fiber reinforced thermoplastic com-
filled epoxy composites, Eng. Fract. Mech. 160 (2016) 22–41.
posites using fused deposition modeling, Compos. Part B Eng. 80 (2015) 369–378.
[28] C. Garzon, et al., Effect of carbon-based particles on the mechanical behavior of iso-
[8] H.L. Tekinalp, et al., Highly oriented carbon fiber–polymer composites via additive
tactic poly(propylene)s, Macromol. Mater. Eng. 301 (4) (2016) 429–440.
manufacturing, Compos. Sci. Technol. 105 (2014) 144–150.
[29] B.D. Le, et al., Modeling damages and cracks growth in composite with a 3D discrete
[9] O.S. Carneiro, A.F. Silva, R. Gomes, Fused deposition modeling with polypropylene,
element method, Compos. Part B Eng. 91 (2016) 615–630.
Mater. Des. 83 (2015) 768–776.
[30] P.W. Harper, S.R. Hallett, Cohesive zone length in numerical simulations of compos-
[10] C. Casavola, et al., Orthotropic mechanical properties of fused deposition modelling
ite delamination, Eng. Fract. Mech. 75 (16) (2008) 4774–4792.
parts described by classical laminate theory, Mater. Des. 90 (2016) 453–458.
[31] W.-G. Jiang, et al., A concise interface constitutive law for analysis of delamination
[11] T.-M. Wang, J.-T. Xi, Y. Jin, A model research for prototype warp deformation in the
and splitting in composite materials and its application to scaled notched tensile
FDM process, Int. J. Adv. Manuf. Technol. 33 (11−12) (2006) 1087–1096.
specimens, Int. J. Numer. Methods Eng. 69 (9) (2007) 1982–1995.
[12] Q. Sun, G.M.R., C.T. Bellehumeur, P. Gu, Effect of processing conditions on the bond-
[32] M. Faker, et al., Rheology, morphology and mechanical properties of polyethylene/
ing quality of FDM polymer filaments, Rapid Prototyp. J. 14 (2) (2008) 72–80.
ethylene vinyl acetate copolymer (PE/EVA) blends, Eur. Polym. J. 44 (6) (2008)
[13] D. Qiu, N.A. Langrana, Void eliminating toolpath for extrusion-based multi-material
1834–1842.
layered manufacturing, Rapid Prototyp. J. 8 (1) (2002) 38–45.
[33] D. Notta-Cuvier, et al., Damage of short-fibre reinforced materials with anisotropy
[14] K. Chou, Y. Zhang, A parametric study of part distortions in fused deposition model-
induced by complex fibres orientations, Mech. Mater. 68 (2014) 193–206.
ling using three-dimensional finite element analysis, Proc. Inst. Mech. Eng. B J. Eng.
[34] T. Kuboki, et al., Mechanical properties and foaming behavior of cellulose fiber rein-
Manuf. 222 (8) (2008) 959–968.
forced high-density polyethylene composites, Polym. Eng. Sci. 49 (11) (2009)
[15] A. Kantaros, D. Karalekas, Fiber Bragg grating based investigation of residual strains
2179–2188.
in ABS parts fabricated by fused deposition modeling process, Mater. Des. 50 (2013)
44–50.