You are on page 1of 9

Plastics, Rubber and Composites

Macromolecular Engineering

ISSN: 1465-8011 (Print) 1743-2898 (Online) Journal homepage: http://www.tandfonline.com/loi/yprc20

Effect of the printing bed temperature on the


adhesion of parts produced by fused filament
fabrication

Martin Spoerk, Joamin Gonzalez-Gutierrez, Janak Sapkota, Stephan


Schuschnigg & Clemens Holzer

To cite this article: Martin Spoerk, Joamin Gonzalez-Gutierrez, Janak Sapkota, Stephan
Schuschnigg & Clemens Holzer (2018) Effect of the printing bed temperature on the adhesion of
parts produced by fused filament fabrication, Plastics, Rubber and Composites, 47:1, 17-24, DOI:
10.1080/14658011.2017.1399531

To link to this article: https://doi.org/10.1080/14658011.2017.1399531

© 2017 The Author(s). Published by Informa


UK Limited, trading as Taylor & Francis
Group

Published online: 08 Nov 2017.

Submit your article to this journal

Article views: 3170

View Crossmark data

Full Terms & Conditions of access and use can be found at


http://www.tandfonline.com/action/journalInformation?journalCode=yprc20
PLASTICS, RUBBER AND COMPOSITES, 2018
VOL. 47, NO. 1, 17–24
https://doi.org/10.1080/14658011.2017.1399531

Effect of the printing bed temperature on the adhesion of parts produced by


fused filament fabrication
Martin Spoerk , Joamin Gonzalez-Gutierrez , Janak Sapkota , Stephan Schuschnigg
and Clemens Holzer
Polymer Processing, Montanuniversitaet Leoben, Leoben, Austria

ABSTRACT ARTICLE HISTORY


For parts produced by fused filament fabrication (FFF) the adhesion between the first printed Received 28 April 2017
layer and the printing bed is crucial, since it provides the foundation to the subsequent Revised 1 September 2017
layers. Inadequate adhesion can result in poor printing quality or destroyed bed surfaces. Accepted 28 October 2017
This study aims at understanding and optimising the adhesion process for parts produced by
KEYWORDS
FFF. The consequences of varying printing bed temperatures on the adhesion of two Additive manufacturing;
commonly used printing materials on two standard bed surfaces were investigated by means fused filament fabrication;
of an in-house built adhesion measurement device and complemented by contact angle adhesion; polylactic acid;
measurements. This study shows a significant increase in adhesion forces, when printing acrylonitrile butadiene
parts at a bed temperature slightly above the glass transition temperature of the printing styrene; contact angle
material. measurements; morphology;
glass transition temperature

Introduction
tapes like polyimide (PI) or blue painters tape; (iv) to
Fused filament fabrication (FFF) or fused deposition slightly increase the tape surface roughness by sanding
modelling (FDM) is an extrusion-based additive man- it; (v) to apply water-soluble glues, hair sprays or
ufacturing technique, which relies on the extrusion of special coatings; (vi) to print on a plate or film of the
thermoplastic filaments to produce a three-dimen- same material that is being printed; (vii) to print on
sional object in a layer-by-layer manner [1]. The cleated surfaces; and (viii) to increase the temperature
adhesion of the first printed layer onto the printing of the printing bed to a recommended value for a given
bed is critical, as without proper adhesion the final material [2,3,5–7].
part cannot be built [2]. Inadequate adhesion results Methods (i–ii) are known prerequisites done before
in poor-quality printed objects, likely due to shifts, every printing. Methods (iii–vii) require the appli-
warps or delaminations of the object during the print- cation of additional materials on the printing bed,
ing process [3]. Therefore, the adhesion between the which may be difficult to apply. Based on our experi-
extruded polymer and the printing bed should be ence, the risk of uneven printing bed surfaces caused
high enough to keep the printed object in place during by overlapping seams, folds, creases or air bubbles, is
printing. On the other hand, after printing the drastically increased when using methods (iii–iv). In
adhesion should be low enough so that the part can this context, one practical solution is to increase the
be removed easily from the printing bed without dama- bed temperature to improve the adhesion of the
ging the produced part and the bed surface [4]. printed material onto the printing bed during print-
FFF printers are commonly composed of printing ing. However, directly after printing, a non-destruc-
beds made of glass or polymers [2]. The printed objects tive removal of the printed part may not be
are supposed to adhere consistently onto these sur- attainable at this bed temperature. If the occurrence
faces. However, this is not always the case. Thus, to of welding can be precluded, it may be favourable
improve the adhesion of the first layer to the printing in this regard to cool the printing bed first to a certain
bed, several things are recommended: (i) to clean the temperature, at which the adhesion forces are suffi-
printing surface to remove grease and residues from ciently reduced. Two main problems arise. The first
the bed; (ii) to level the printing bed so that the first problem is to determine this operational temperature
layer is in close contact with the printing bed; (iii) to range, whose upper limit, effective during the print,
cover the printing surface with polymeric films or ensures sufficient adhesion during printing, and

CONTACT Martin Spoerk martin.spoerk@unileoben.ac.at Polymer Processing, Montanuniversitaet Leoben, Otto Gloeckel-Straße 2, Leoben 8700,
Austria
© 2017 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-
nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or
built upon in any way.
18 M. SPOERK ET AL.

whose lower limit, adjusted after the completion of diameter was used for the production of the adhesion
the print, provides little adhesion for the removal of test specimens (see section ‘Adhesion measurement’).
the part. Second, the determination of the best A constant extrudate flow rate of 2 mm3 s−1 and a
material combination of the printing bed and the fila- printing speed of 50 mm min−1 were kept constant
ment is critical, e.g. to avoid excessive welding, for all specimens. As the adhesion of printed parts is
especially when using novel materials. dependent on the first layer height [5], the layer thick-
So far, no study revealed any strategies on identify- ness of the first layer was set constant to 0.2 mm. For
ing an optimal printing bed temperature range for var- PLA and ABS die temperatures of 220 and 255°C
ious materials and its effect on the adhesion of samples were used, respectively. The bed temperature was var-
produced by FFF in a systematic manner. For this ied between 30 and 120°C (measured with reference to
reason, the present work attempts to close this gap by the metal surface below the printing bed) with a step of
evaluating the adhesion forces by means of an in- 10 K. The heaters for the printing bed are located
house developed device that measures the forces underneath the metal surface below the printing bed
required to shear-off printed strands from printing sur- and they were controlled by the pre-installed heating
faces at different temperatures. Using this device, the system of the Hage 3DpA2.
effect of temperature on the adhesion of two com- For each material at a given set of conditions, 16
monly used printing materials, namely acrylonitrile- strands were printed, in which each strand was
butadiene-styrene (ABS) and polylactic acid (PLA) 100 mm in length and consisted of three layers, result-
onto two different printing surfaces (glass and PI) ing in a total height of 0.6 mm. The width of each layer
was investigated. Moreover, these measurements were is in the range of 1.65 and 1.9 mm and is independent
complemented by surface tension measurements and of the bed temperature. The distance between adjacent
morphology analyses of the topography of the contact strands was 12 mm. In order to obtain well repeatable
area. The systematic approach in the present work results, the printing bed was levelled perfectly before
illustrates an efficient way to optimise the adhesion each test and the distance between the nozzle and the
of different printing materials on various printing printing bed was checked to be constant at every
bed materials. Especially for novel material com- point on the bed before starting a print.
pounds, this methodology is of great importance for
a reliable printing process.
Adhesion measurement
The adhesion forces between the 3D-printed strands
Experimental
and two printing bed materials were measured by
Materials means of a self-developed shear-off force testing device
(Figure 1(a)) directly mounted on the printing bed. For
PLA and ABS filaments with a mean diameter of
each measurement series, the printing bed was pre-
1.75 mm were used as printing materials. A glass mir-
heated for at least 30 min. Before the print was started,
ror and a PI film, glued to the glass, were used as print-
the possibly aged material in the die was removed, and
ing bed materials. All the materials were supplied by
the printing bed was cleaned residue-free with the
Prirevo e.U., Austria, and used as received. Unless sta-
cleaning agent Arecal (Reca, Austria). After the com-
ted otherwise, their most important properties are
pletion of the print of 16 strands (see section ‘Printer
summarised according to the supplier’s data sheet in
settings’ for the detailed settings), the bed temperature
Table 1.
was kept constant by the heaters under the building
platform and the regulation of the printing bed. The
Printer settings shear-off force testing device was fixed on each corner
of the printing bed by means of clamps to prevent any
A Hage 3DpA2 (Hage Sondermaschinenbau GmbH &
relative movement of the device, while shearing off the
Co. KG, Austria) with a brass nozzle of 0.5 mm in
printed strands. The testing device was pre-heated for
5 min on the printer in order to prevent warpage,
Table 1. The glass transition temperature TG, the density ρ, the which can critically influence the measurements.
specific heat capacity at constant pressure CP and the thermal Before the measurement started, the metallic shearing
conductivity λ of the materials investigated.
block ((2) in Figure 1(a)) was adjusted depending on
Materials
the printing bed material so that the vertical distance
Properties PLA ABS Glass PI
of 0.1 mm between the shearing block and the printing
TG a (°C) 60.6 110.1 … 348.6
ρ (kg m−3) 1250 1004 2200 … bed remained constant for the whole measurement
CP (J kg−1 K−1) 1900 1800 750 … area. Each strand was tested separately. For each
λ (W m−1 K−1) 0.180 0.127 1.400 …
measurement, the metallic shearing block moved hori-
a
TG was characterised by means of differential scanning calorimetry on a
Mettler Toledo DSC 1 equipped with a gas controller GC 200 (Mettler zontally over the printing bed at a constant speed of
Toledo GmbH, Switzerland). 2 mm s−1 and sheared-off one printed strand. The
PLASTICS, RUBBER AND COMPOSITES 19

Figure 1. Schematic of the adhesion force testing device (a) and an illustration of a typical test sequence by means of a graph
representing the measured adhesion forces as a function of the displacement and a sketch of the corresponding measured strands
on the printing bed (b). In (a) the components of the measuring device are labelled: (0) printing bed, (1) gliding frame, (2) metallic
shearing block with linear bearings, (3) force transducer, (4) stabilising bar, (5) displacement screw, (6) transmission gear, (7) dis-
placement transducer, (8) motor and (9) printed strands. In (b) the geometry of the shearing block (rectangular element with the
horizontal arrow), the strands and the printing bed are shown schematically in a lateral and top view along with the most important
dimensions.

tested strand was then manually removed from the at different temperatures. Prior to the surface analysis,
printing bed before the subsequent strand was tested. ABS and PLA were pressed to plates (160 × 160 ×
At the beginning of each measurement, the shearing 2 mm3) in a Collin P200PV vacuum press (Dr. Collin
block was retracted to guarantee that a lateral distance GmbH, Germany) at 200°C for PLA and 270°C for
of 12 mm between the shearing block and the next ABS, and 150 bar for 25 min. As printing beds, the
strand was preserved to ensure similar load cell con- glass mirror and the PI film were investigated. The
ditions in every measurement. The forces, measured resulting contact angle between each material and the
by a miniature load cell (U9C 1 kN, Hottinger Baldwin test-liquids 1-Bromonaphthalene (CAS 90-11-9) and
Messtechnik GmbH, Germany) ((3) in Figure 1(a)), Ethylene carbonate (CAS 96-49-1) was measured. Fif-
were displayed as a function of the displacement teen repetitions were performed per combination, in
(Figure 1(b)), which was measured by an inductive dis- which the propagation of uncertainty was considered.
placement transducer (W100, Hottinger Baldwin Mes- In all graphs, the corresponding error is displayed.
stechnik GmbH, Germany) ((7) in Figure 1(a)). The The results were evaluated as described in [8,9]. For
analog data from the force and displacement sensors the contact angle measurements at higher tempera-
were collected at 300 Hz with a Spider 8 Data Acqui- tures, the contact temperature between the printing
sition System (Hottinger Baldwin Messtechnik die and the printing bed was calculated as:
GmbH, Germany) and the software CatmanAP  
V3.5.1 (Hottinger Baldwin Messtechnik GmbH, Aus- TF · lF · rF · CP,F + TB · lB · rB · CP,B
tria). A zero-force baseline was set for each measure- TContact =   ,
lF · rF · CP,F · lB · rB · CP,B
ment. For each setting (section ‘Printer settings’), the
(1)
force maxima of the 16 tested strands were evaluated
to a significance level of 5%. in which TContact is the contact temperature, TF and TB
are the filament and printing bed temperatures, λF and
λB are the thermal conductivities of the filament and
Contact angle measurements
the bed material, ρF and ρB are the densities of the fila-
Contact angle measurements were performed with a ment and the bed material, and CP,F and CP,B are the
Krüss DSA100 (Krüss GmbH, Hamburg, Germany) specific heat capacities at constant pressure of the
20 M. SPOERK ET AL.

filament and the bed material. The obtained values is increased, segments of the polymer chains can dif-
necessary for the calculation of TContact are summarised fuse into the interface and the amount of adhesion is
in Table 1. For TContact of the PI film, the values for λB, dependent on the extent of interdiffusion and chain
ρB and CP,B are taken from the glass mirror, as the PI interpenetration into the interface [13].
film with a thickness of 0.05 mm has an insignificant As the temperature keeps increasing beyond 70°C,
effect on TContact. For further reference, the tempera- the adhesion forces drop independent of the printing
ture refers to the printing bed temperature. bed material to roughly 50–60% of the forces for a
bed temperature of 70°C. The described decrease can
Microscopy be attributed to the changed interaction between the
molten polymer and the printing surface. A higher
The contact surfaces of selected printed strands were negative Laplace pressure is generated, which drags a
analysed in the optical microscope Olympus BX51 large amount of polymer to some parts of the interface,
(Olympus Life Science Europe GmbH, Germany) at a while other parts are underfilled. Such changes in
magnification of 50× under reflected light. The speci- pressure can lead to what is referred as fingering or
mens were used after testing their adhesion forces Saffman–Taylor instabilities, which are wavy undula-
(section ‘Adhesion measurement’) without further tions at the periphery of the contact area that can
modification. decrease the adhesion force [14]. For the PLA printed
onto the glass, the adhesion forces reach a plateau
after 80°C. For the PI film, the behaviour is more com-
Results
plex, since the PI is also a polymeric material and its
Adhesion forces segment mobility increases with temperature [15].
Therefore, a continuous increase in the adhesion
The adhesion forces for both PLA (Figure 2(a)) and
force is observed as the temperature increases from
ABS (Figure 2(b)) show a strong dependence on the
80 to 100°C. Measurements beyond 100°C were not
printing bed temperature. In the case of PLA printed
possible, because the adhesive used to glue the PI
on both bed materials (Figure 2(a)), up to a bed temp-
film onto the glass failed at these high temperatures
erature of 60°C a steady rise in the adhesion force is
and measured forces.
observed. This increase can be attributed to an
The adhesion behaviour of ABS printed onto the
enhanced chain mobility of the deposited filament
glass or PI film exhibits a similar trend to that of PLA.
with higher temperatures [10,11]. A point to highlight
The adhesion forces increase with rising temperature
is the strong increase in the adhesion force as the temp-
in a non-linear fashion with a maximum slightly
erature of the printed bed augments from 60 to 70°C.
above the TG. As it was not possible to measure at temp-
For PLA printed on PI the adhesion forces rise from
eratures higher than 120°C due to limitations of the
51 ± 8 to 322 ± 47 N, whereas the measured forces on
printer bed heaters, a peak could not be observed in
glass are enhanced even more, from 73 ± 19 to 651 ±
the adhesion force diagram for ABS. However, it is
17 N. This is related to the glass transition temperature
expected that there would be a peak after the TG, since
(TG) of the printing material (60.6°C for PLA). Around
this behaviour was also observed for other amorphous
the TG, the segmental mobility of macromolecules is
polymers such as poly(methyl methacrylate) [14]. The
the highest for a material, which might result in
adhesion between ABS and PI is much better than
enhanced adhesion between polymeric surfaces and
that of ABS and the glass, due to closer similarities of
other materials [10,12]. When the segmental mobility
the polar and disperse fractions of the two materials,
expressed by the lower interfacial tensions, which is
shown in section ‘Contact angle measurements’. When
comparing the adhesion force values for PLA and
ABS, it can be clearly seen that those obtained by ABS
are significantly lower than those of PLA. However,
both materials could be printed without the risk of
detaching from the printing bed. For large-area parts,
the warpage needs to be considered, as it counteracts
the adhesion due to uneven shrinkage of the printed
layers that tend to pull the part away from the printing
surface [16]. From the authors’ experience, adhesion
forces of at least 200 N, as obtained from adhesion
measurements on strands by means of the shear-off test-
Figure 2. Adhesion forces as a function of printing bed temp- ing device, are sufficient for larger parts to hold them in
erature and bed material (glass or PI) for PLA (a) and ABS (b). place during printing. Hence, for ABS a bed temperature
The TGs of the filament materials are depicted as the blue lines. of at least 120°C (T > TG) and the bed material PI are
PLASTICS, RUBBER AND COMPOSITES 21

necessary to ensure proper adhesion during printing, tension from contact angle measurements [17]. The
whereas for PLA bed temperatures higher than 60°C adhesion between two contact materials is inversely pro-
(T > TG) independently of the bed material are more portional to the interfacial tension values, and such
than sufficient for a successful printing. On the contrary, values are inversely dependent on the surface energies
for very high adhesion forces, such as occur e.g. for PLA and polarities of the materials in contact [9].
and glass at 70°C, a second issue has to be considered. Figure 3 compares the interfacial tension to the
Too high adhesion forces can lead to a damage of the measured adhesion force at two temperatures for both
printing bed during cooling. Therefore, in fact, the opti- printing materials and printing beds. The selected temp-
mal printing bed temperature has to be carefully selected eratures are 10 K below and above the TG of each printed
within a moderate adhesion force range. For PLA and material. Despite the big variations, a trend towards
glass, for example, printing bed temperatures between decreased interfacial tensions for increasing temperatures
80 and 120°C can be recommended. can be seen for both PLA and ABS (Figure 3(a,c)). This
As can be seen from Figure 2, for PLA a temperature trend is in accordance with the adhesion force results
close to room temperature can be recommended for a (Figure 3(b,d)), in which a drastic increase is observed
damage-free removal after printing for both surface for higher temperatures. Moreover, the trend towards
materials, as the adhesion force is close to zero at this higher adhesion forces for printing PLA on the glass
temperature. However, for ABS the adhesion to PI than on the PI surface is reflected by the interfacial tension
and glass is negligible already at 50°C. Hence, the fin- values (Figure 3(a)). Also for ABS, the interfacial tension
ished parts can be easily removed also at a slightly elev- results (Figure 3(c)) reflect the trend of the adhesion
ated temperature. measurements (Figure 3(d)), in which the adhesion on
the glass is inferior to that on the PI film. However, the
overall trend in the contact angle measurements (Figure
Contact angle measurements
3(a,c)) does not reflect the significant difference in the
The adhesion force is dependent on the compatibility adhesion forces (Figure 3(b,d)), but the contact angle
between the two surfaces [13], which is affected by the measurements rather exhibit non-significant differences
polarity and the thermodynamic characteristics of the in the interfacial tension. Hence, the adhesion mechanism
two interacting surfaces. One way to determine the com- cannot be solely explained by surface chemical analyses,
patibility between surfaces is to calculate the interfacial such as contact angle measurements.

Figure 3. Interfacial tension obtained from contact angle measurements as a function of printing bed temperatures for PLA (a) and
ABS (c) compared to adhesion forces of PLA (b) and ABS (d) at the same temperatures.
22 M. SPOERK ET AL.

Figure 4. Optical microscopy images of printed PLA parts after being sheared-off from different surfaces at different temperatures:
(a) glass at 50°C, (b) glass at 70°C, (c) PI at 50°C and (d) PI at 70°C. In the top right corner of each image the corresponding adhesion
force is shown. In the bottom left corner of each image one representative force–displacement diagram with the same displace-
ment labelling is shown to visualise the difference in their shear failure. The circles highlight sheared-off pieces of the printing bed
and the arrows the loading direction of the measuring device.

Microscopy analysis
Another indication for the strong adhesion of the
The adhesion of polymers onto different surfaces is specimen printed at 70°C (Figure 4(b)) is the defor-
additionally affected by the surface topology and the mation of the first layer, displayed in the bottom of
impurities that may be present on the surface [18]. Figure 4(b). This originates from the local bending
The magnified surfaces of the printed parts after due to its strong adhesion and the shearing block.
being sheared-off at different temperatures as well as The white circles, highlighted in Figure 4(b), represent
one representative force–displacement curve each are the presence of pieces of glass that had been sheared-off
shown in Figure 4 exemplarily for PLA. Similar results from the glass during the adhesion force measure-
are expected for ABS and hence they were not included ments. This finding confirms that extreme adhesion
in this manuscript. It can be seen that the temperature between the printed part and the printing bed might
does indeed have an effect on the resulting topology of not be favourable, since it can destroy both material
the printed parts. in contact, as discussed in section ‘Adhesion forces’.
PLA printed on the glass at 50°C (Figure 4(a)) shows When the surfaces of printed PLA on the PI film are
some voids on the surface, whereas the part printed at compared to those on the glass, an opposite trend is
70°C (Figure 4(b)) has a slightly smoother surface. The observed: The smoother surface of the specimen
higher amount of voids reduces the effective contact printed at 50°C (Figure 4(c)) has a lower adhesion
area, resulting in adhesion forces of only 55 ± 16 N. force (70 ± 11 N) than that of the specimen printed at
On the other hand, the smoother surface obtained by 70°C (Figure 4(d)) (320 ± 47 N). However, when print-
printing on the glass at 70°C indicates a better contact ing on PI, the rougher surface does not exhibit voids,
between the PLA and the glass. Due to the increased but rather channels oriented parallel to the length of
diffusion of the polymer onto the glass and the reduced the printed part.
interfacial tension, this leads to a higher adhesion force The different behaviour of the specimens printed on
of 650 ± 17 N and to a considerably broader adhesion PI might be related to the fact that the PI is indeed a
force peak as can be seen in the insert in Figure 4(b). foil, which is in turn glued on another printing bed
PLASTICS, RUBBER AND COMPOSITES 23

layer, while the glass is itself a rigid plate. At high temp- ORCID
eratures, when the adhesion force between PLA and PI Martin Spoerk http://orcid.org/0000-0001-5606-3607
is high, it can be speculated that the interface between Joamin Gonzalez-Gutierrez http://orcid.org/0000-0003-
the PI foil and the lower printing bed layer can pose a 4737-9823
weak link. As can be seen in Figure 4(d), the PI foil is Janak Sapkota http://orcid.org/0000-0003-2296-5623
not torn off in the course of the shearing experiments, Stephan Schuschnigg http://orcid.org/0000-0002-2601-
967X
indeed it is not damaged. The shearing, which in the
Clemens Holzer http://orcid.org/0000-0001-5149-7895
case of the PI foil does not take place as a uniform
motion and thus causes the channels, can also be cor-
related with the force–displacement curve (insert
Figure 4(d)), which does not exhibit one peak, but sev- References
eral local maxima and inflexion points. [1] Gibson I, Rosen DW, Stucker B. Additive manufactur-
These channels can increase the contact area between ing technologies: 3D printing, rapid prototyping, and
the deposited filament and the flexible PI film. Thus, the direct digital manufacturing. 2nd ed. New York
surface favours more diffusion of the PLA molecular (NY): Springer; 2015.
[2] Carneiro OS, Silva AF, Gomes R. Fused deposition
segments, leading to higher adhesion forces. Also for modeling with polypropylene. Mater Des.
the surface represented in Figure 4(d), the local defor- 2015;83:768–776.
mation in the bottom of the illustration is observed [3] Shafer CS, Siddel DH, Merriman AL, et al. Cleated
and a piece of the bed surface is sheared-off during print surface for fused deposition modeling. In:
the measurement (highlighted by a white circle). Bourell DL, Crawford RH, Seepersad CC, Beaman JJ,
Fish S, Marcus H, editors. Proceedings of the 26th
Annual International Solid Freeform Fabrication
Conclusion Symposium – An Additive Manufacturing
Conference. Austin, TX: The University of Texas at
We present here a systematic study on the effect of the Austin; 2016. p. 1359–1365.
printing bed temperature on the adhesion of parts [4] Gonzalez-Gutierrez J, Duretek I, Holzer C, et al. Filler
content and properties of highly-filled filaments for
printed by FFF by means of an in-house developed
fused filament fabrication of magnets. In: ANTEC
adhesion measuring device. It was observed exemplarily Anaheim 2017: The Plastics Technology Conference;
for PLA and ABS that the optimal adhesion of the 2017.
printed sample to the printing bed can be achieved by [5] Hashemi Sanatgar R, Campagne C, Nierstrasz V.
heating the printing bed at temperatures slightly above Investigation of the adhesion properties of direct 3D
the TG of the filament material. Increasing the tempera- printing of polymers and nanocomposites on textiles:
effect of FDM printing process parameters. Appl Surf
ture above the filament’s TG leads to a reduction of the Sci. 2017;403:551–563.
surface tension between the printing bed and the print- [6] Gaal G, Mendes M, de Almeida TP, et al. Simplified
ing material and to a larger contact area that ultimately fabrication of integrated microfluidic devices using
causes better adhesion between the bed and the filament. fused deposition modeling 3D printing. Sens Actuat
We believe that the process presented in this study is B Chem. 2017;242:35–40.
[7] Micro Electronics Inc. Fused filament fabrication
readily applicable to other thermoplastic polymers
(FFF) material properties & considerations. [Cited
with a TG above room temperature. Our findings pro- 2017 Apr 3]. Available from: http://www.powerspec.
vide the first steps for finding optimal printing bed com/support/3DPrinters/3DPrintingTips.pdf
temperatures for novel material compounds. [8] Berger GR, Steffel C, Friesenbichler W. A study on the
role of wetting parameters on friction in injection
moulding. Int J Mater Prod Technol. 2016;52(1/2):193.
Acknowledgements [9] Owens DK, Wendt RC. Estimation of the surface free
energy of polymers. J Appl Polym Sci. 1969;13
Special thanks go to Petra Erdely for fruitful discussions, Phi- (8):1741–1747.
lipp Huber for the construction of the adhesion device and [10] Vinogradov GV, Bartenev GM, Elkin AI, et al. Effect of
Florian Arbeiter, Gerald Berger, Radoslav Guran and Chris- temperature on friction and adhesion of crystalline
tof Lichal for their help with the measurements. polymers. Wear. 1970;16:213–219.
[11] Forrest JA, Dalnoki-Veress K, Dutcher JR. Interface
and chain confinement effects on the glass transition
Disclosure statement temperature of thin polymer films. Phys Rev E.
No potential conflict of interest was reported by the authors. 1997;56(5):5705–5716.
[12] Zeng H, Maeda N, Chen N, et al. Adhesion and friction
of polystyrene surfaces around Tg. Macromolecules.
Funding 2006;39(6):2350–2363.
[13] Jabbari E, Peppas NA. Polymer-polymer interdiffusion
This work was supported by the Austrian Research Pro- and adhesion. J Macromol Sci C: Polym Rev. 1994;34
motion Agency (Österreichische Forschungsförderungsge- (2):205–241.
sellschaft (FFG)) as part of the AddManu project under [14] Kim K-S, Heo J-C, Kim K-W. Effects of tempera-
grant 849297. ture on the microscale adhesion behavior of
24 M. SPOERK ET AL.

thermoplastic polymer film. Tribol Lett. 2010;38 polypropylene composites in extrusion-based additive
(2):97–106. manufacturing. Macromol Mater Eng. 2017;1700143.
[15] Loos J, Tian M, Rastogi S, et al. An investigation on [17] Clint JH. Adhesion and components of solid surface
chain mobility in solid state polymer systems. J energies. Curr Opin Colloid Interface Sci. 2001;6
Mater Sci. 2000;35(20):5147–5156. (1):28–33.
[16] Spoerk M, Sapkota J, Weingrill G, et al. Shrinkage [18] Awaja F, Gilbert M, Kelly G, et al. Adhesion of poly-
and warpage optimization of expanded-perlite-filled mers. Prog Polym Sci. 2009;34(9):948–968.