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

net/publication/300885109

Analysis of the resin transfer molding (RTM) process for FRP and its process
simulation fundamentals

Conference Paper · May 2015

DOI: 10.1063/1.4918428

CITATIONS READS
4 954

2 authors, including:

Stefan Caba
Technische Universität Ilmenau
4 PUBLICATIONS   9 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Advanced Process Design for RTM View project

All content following this page was uploaded by Stefan Caba on 22 July 2016.

The user has requested enhancement of the downloaded file.

 Analysis of the resin transfer molding (RTM) process for FRP and its process
simulation fundamentals
S. Caba and M. Koch

Citation: AIP Conference Proceedings 1664, 060010 (2015); doi: 10.1063/1.4918428

View online: http://dx.doi.org/10.1063/1.4918428
View Table of Contents: http://scitation.aip.org/content/aip/proceeding/aipcp/1664?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in

Analysis of Molding Process for Epoxy Resin Used for Electrical Insulator
AIP Conf. Proc. 908, 1193 (2007); 10.1063/1.2740972

Meshless methods with application to Resin Transfer Molding simulation

AIP Conf. Proc. 907, 1378 (2007); 10.1063/1.2729707

Resin Flow Analysis in the Injection Cycle of a Resin Transfer Molded Radome
AIP Conf. Proc. 907, 951 (2007); 10.1063/1.2729636

A Parallel Computational Environment for Modeling the Resin Transfer Molding Process
AIP Conf. Proc. 712, 305 (2004); 10.1063/1.1766541

Fiber optic sensors for use in monitoring flow front in vacuum resin transfer molding processes
Rev. Sci. Instrum. 68, 2156 (1997); 10.1063/1.1148065

Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 191.96.247.167 On: Fri, 22 Jul 2016 01:19:48
 Analysis Of The Resin Transfer Molding (RTM) Process For
FRP And Its Process Simulation Fundamentals
S. Cabaa and M. Kocha
a
Technische Universität Ilmenau, Fachgebiet Kunststofftechnik, Germany – [email protected]

Abstract. The industrialization of the resin transfer molding process is a major objective in reducing production cost for
FRPs thus making it more competitive for use in large scale applications. Normally, shorter cycle times run into the risk
of quality inconsistencies. The creation of voids is influenced by various process parameters, such as mold temperature,
resin flow rate and fiber volume content. With knowledge of the effects of these parameters on the process, a specific
approach to reducing void content is made while reducing processing time. In order to achieve this objective, in
particular, capillary forces and its influence on impregnation of the fiber bundle were examined. A DOE approach was
chosen to lead to the determination of interactions. The results target at a pre-determination of required process parameter
to obtain consistent part quality with reduced cycle time.
Keywords: Resin Transfer Molding, Composite Manufacturing, Capillary Number, Void Reduction
PACS: 81.05.Qk

INTRODUCTION
Lightweight design is one of the most innovative sectors in engineering. It is driven by high demand from
aircraft and automotive industries. In particular, the use of fiber reinforced plastics (FRP) is considered to save a
considerable amount of carbon dioxide emissions by reducing the weight of vehicles. Beyond new methods of
design and construction as well as methods of calculating the manufacturing process for these materials, there is still
considerable need to understand process fundamentals and conduct detailed experimentation. The resin transfer
molding process (RTM) has been identified to enable the production of parts with high fiber reinforcement content
with smooth surfaces in short cycle times [1]. Current processing still suffers from high void content and often times
do not achieve expected results for large production volumes. The automotive industry requires output rates of more
than ten parts per hour [2] ranging up to multiple hundreds per day. In order to be able to achieve such rates,
processing must be understood much better to correlate factors impacting void creation and cycle.

VOID FORMATION
There are different mechanisms of void formation in the RTM-process. Voids can be divided in two classes –
spherical and cylindrical pores. Spherical voids are typically located between fiber strands while cylindrical voids
occur within rovings between individual filaments [3]. The mechanisms creating these patterns are characterized as
chemical and physical processes as shown in figure 1. Chemical mechanisms result from humidity, reaction
byproducts and outgassing contaminations and create particularly spherical voids. Thus, the application of pure
input materials as well as an attentive material handling while placing fibers into a mold can reduce the content of
spherical pores. Contaminations and humidity can also be removed by outgassing the reacting agents of the resin
and drying the reinforcements. Physical void formation results from the flow pattern, which is influenced by the
local permeability. The fibers in the strands follow a statistical distribution. As a consequence, the permeability
respectively the flow pattern may vary considerably subject to the fiber volume content [4]. The resin always takes
the line of the least flow resistance. This may effect in a flow around the areas with high fiber volume content.
Cylindrical pores in the strand are created.

Proceedings of PPS-30
AIP Conf. Proc. 1664, 060010-1–060010-5; doi: 10.1063/1.4918428
© 2015 AIP Publishing LLC 978-0-7354-1309-2/$30.00

060010-1
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 191.96.247.167 On: Fri, 22 Jul 2016 01:19:48
 humidity

reaction
chemical byproducts

void
contaminations

stastistical fiber
distribution
cylindrical physical
unbalanced
spherical flow front

FIGURE 1 Void characterization and formation mechanisms

The predominantly prevailing mechanism results from the differences in permeability and a variability of
capillary forces between fiber strands and the space between individual rovings. The permeability is much higher in
the strand because of much smaller flow channels between individual filaments. On the other hand, the wicking
effects caused by capillary forces profit from the high surface to volume ratio in the strands. This may lead to an
unbalanced flow front. Figure 2 demonstrates the mechanisms of void formation by unbalanced flow fronts. Once
the resin flow velocity in the bundle is higher, spherical bubbles will be formed. A higher flow rate between rovings
free flow volume result in cylindrical voids.

FIGURE 2 Void formation by unbalanced flow fronts

SIMULATION OF VOID FORMATION VIA THE CAPILLARY NUMBER

As shown above most voids emerge from flow phenomena. The flow in the mold can be characterized by the
capillary number Ca* as defined in equation (1). This dimensionless key figure indicates the void type and the
expected void content [5]. This number is a ratio of the mainly pressure induced flow between the bundles and the
microscopic flow in the bundles. The microscopic flow depends on the capillary pressure. The difference of the
surface tension between fiber surface and resin and their interaction are the driving power behind the capillary
pressure.
ఎή௩ത
‫= כܽܥ‬ (1)
ఊήୡ୭ୱ ఏ

With: Ș UHVLQYLVFRVLW\vത = flow front velocity Ȗ UHVLQVXUIDFHWension ș FRQWDFWDQJOH

060010-2
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 191.96.247.167 On: Fri, 22 Jul 2016 01:19:48
 Capillary numbers less than 0.002 indicate the potential creation of spherical voids. This means that the overall
resin velocity is lower than the wicking effect in the bundles. A value exceeding 0.03 marks the formation of
microscopic voids in the bundle [3].

The calculation of the capillary number requires the knowledge of all four variables. This knowledge enables the
predetermination of void formation as a result of process conditions and, therefore, a means to avoid it. The factors
are determined by measurements of material properties in combination with the RTM process.

The surface tension can be measured using different methods. Epoxy resins show typical surface tension of
45 mN/m to 48 mN/m calculated by the drop weight. The measurement of the contact angle is way more difficult.
The challenge is the capillary effect itself. There is almost no steady state as the resin is drawn into the fiber bundle
and the contact angle is time dependent. Literature only describes few values and reproduceable data is hard to
obtain. Measurements were conducted by means of the Wilhelmy method with values of 34° ±5° [6]. In this method
the fibers are dipped into a drop of the resin. The angle between the fiber and the resin is measured. Own
measurements were conducted and confirmed the results with a contact angle of 32.5° ±7° [7].

FIGURE 3 Measurement of the contact angle between resin and fiber via the Wilhelmy method [7]

The flow front velocity is a function of the pressure gradient, the permeability, the viscosity and the distance
from the inlet, thus, the flow length. The flow resistance grows with the distance from the inlet. So the velocity
drops along the flow length during the injection time considering an invariant inlet pressure. On the one hand, the
viscosity decreases due to the reaction temperature increase of the resin within the mold. On the other hand it
increases subject to reaction propagation during the injection time. Thus, the flow front velocity is impacted by both
features. This change in viscosity highly impacts on the capillary number that determines the creation of voids. The
complexity of void creation therefore is a result of the changes in the capillary number during the injection process.

In an optimized RTM process the closed mold time is preferably short, the injection time is to be short and the
resin should cure fast. However, a fast cure results in an increase of viscosity. It has been assumed as to be steady in
different studies [3][8][9]. Here, we take a more detailed approach to the change of viscosity allowing for a better
understanding of the physical behavior of flow front and void generation. At first, the change in viscosity is
examined using a new approach. An experimental examination of the resin viscosity was done in a plate-plate
rheometer with a diameter of 15 mm using the oscillation method with a gap of 1 mm, a shear of 10 % and 1 Hz.
The resin was measured right after a mixing time of 1 minute with a temperature of 40°C. Further experiments will
be done at the same temperature for comparison reasons.

As expected, the viscosity develops exponentially. A new approach must be able to describe this behavior. The
cumulative function of the Weibull distribution has been chosen and modified to equation (2). The function was
approximated by using the method of the least squares and the Nelder-Mead method for minimizing the difference.

060010-3
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 191.96.247.167 On: Fri, 22 Jul 2016 01:19:48
 The Viscosity behavior at 40°C is shown in figure 4. The comparison between a measured and a calculated plot
based on the Weibull approach shows a very good fit.
ି(ఒή௧)ೖ
ߟ (‫ߟ = )ݐ‬
௠௔௫ ቀ1 െ ݁ ଴ ቁ+ߟ (2)

Viscosity Development of RIM135/RIMH134

0,60
Viscosity measured
Weibull Approach
Viscosity [Pas]

0,40

Weibull parameters:
0,20
Șmax 1821 Pas
Ș0 130 Pas
Ȝ 5.4·10-4
k 1.836
0,00
0 2 4 6 8 10 12 14 16 18 20
Time [min]
FIGURE 4 Reaction of the resin measured in a plate-plate rheometer and approximation with a Weibull function

EXAMINING THE VOID FORMATION

The formation of the flow front and the resulting voids has been examined in a glass mold. A glass fiber
preform consisting of 7 layers of a twill fabric with 280 g/m² was used. The contact angle and the surface tension of
the epoxy resin RIM135/RIMH134 were measured. The mold was heated up to 40° C. A vacuum of 0.8 bar and an
injection pressure of 2 bar were applied. The flow front was measured using a high definition camera. The resulting
capillary number is shown in figure 5. It was calculated using the measured flow front velocity and the viscosity
from the rheometer experiment. It decreases from a high level at the beginning of the injection process down to the
end of the flow path with an area of least voids [3]. The resulting part was examined with a microscope. Samples
were cut at 30 mm, 50 mm and 70 mm along the observed flow path direction.

The result confirms that the void type can be predetermined by calculating the capillary number. An area with
low void content is observed as a result of the ideal range of the capillary number. In locations where the fibers were
wetted with a high capillary number cylindrical voids were formed. In the area of a lower capillary number,
spherical voids in size of about 0.003 were formed. This was not expected because literature [3][5][8][9] refers to
this area being subject to low void content. The background of these observations should be examined in more
detail.

060010-4
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 191.96.247.167 On: Fri, 22 Jul 2016 01:19:48
 Warp direction Fill direction
1E-01
Capillary number Ca*

Area of least voids

1E-02

1E-03
0 20 40 60 80 100 120 140 160
Distance from inlet [mm]

a) b) c)
FIGURE 5 Plot of calculated capillary numbers along the flow path and resulting void content

CONCLUSION AND PERSPECTIVE

The void formation in the RTM process was examined using the capillary number as an indicator for void
formation. The viscosity was found to be an important factor impacting on the capillary number. A new approach
for the viscosity development was derived using the cumulative function of the Weibull equation. In a RTM process
with a glass mold the capillary number was measured using the measured viscosity development from rheometer
data. Beneath an area of low void content an area of an unexpectedly high void content was found. The void
formation in this area will be examined considering more factors like strand diameter, fabric weight, fiber finish and
the fabric weave. For a better handling and the consideration of interdependencies a design of experiments approach
will be chosen for further experimentation.

REFERENCES
1. R. Lässig et al., Series production of high strength composites, Roland Berger, Munich, 2012
2. M. Arnold and P. Mitschang, “Permeabilität als Schlüsselfaktor für kurze Zykluszeiten“; Kunststoffe 2012/3, pp. 45-48
3. C. H. Park and W. I. Lee, “Modeling void formation and unsaturated flow in liquid composite molding process: a survey and
review”; Journal of Reinforced Plastics & Composites 2011/30, pp. 957-977
4. B. R. Gebart, “Permeability of Unidirectional Reinforcements for RTM”; Journal of Composite Materials 26/1992/8,
pp.1100-1133
5. N. Patel and V. Rohatgi, “Microscale flow behavior and void formation mechanism during impregnation through an
unidirectional stitched fiber glass mat”; Polymer Engineering and Science 35/1995/10, pp. 837-851.
6. H. J. Barraza and M. J. Blakely, “Wetting Behavior of Elastomer modified Glass Fiber”; Langmuir 2001/17, pp. 5288-5296
7. F. Wenzel, “Charakterisierung des Einflusses von Feuchte und Oberflächenspannung bei Fasern auf das
Tränkungsverhalten”, Bachelors thesis, Technische Universität Ilmenau, 2013
8. N. Patel and L.J. Lee, “Modeling of Void Formation and Removal in Liquid Composite Molding, Part 1: Wettability
Analysis”; Polymer Composites 17/1996/1, pp. 96-103
9. J.S.U. Schell, “Investigation of the Influence of Meso-Scale Flow Effects in Liquid Composite Moulding processes on the
Laminate Quality”, Ph.D. Thesis, ETH Zurich, 2007

060010-5
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 191.96.247.167 On: Fri, 22 Jul 2016 01:19:48
View publication stats