Embed Size (px)
Citation preview
Page 1
A FORMULATION STUDY OF LONG FIBER THERMOPLASTIC POLYPROPYLENE (PART 4): THE EFFECT OF MOLDING CHANGES
ON MECHANICAL PROPERTIES OF PP LFT PARTS
Creig Bowland and J. vd Woude PPG Industries, Fiber Glass Science and Technology
Abstract
In part four of this multiyear study the relationship between the Polypropylene Long Fiber Thermoplastics (PP LFT) and the processes used to mold are presented. Parts are molded by injection molding, compression molding and plunger tool using the same formulation in the PP LFT part. The effects of glass fiber diameter, sizing and changes in molding conditions were explored. Prior work showed significant differences between injection molding and compression molding. The use of the plunger tool has given a unique opportunity to look at the effects of molding conditions and fiberglass products on the final part performance.
Background
Long Fiber Thermoplastics have enjoyed a long run of double digit growth and have found general acceptance as structural materials. Polypropylene (PP) LFT materials offer strength and stiffness and are recyclable having a long shelf life. Consequently, their introduction has led to an increased penetration of the automotive market for structural thermoplastic composites. These applications typically are metal replacement and facilitate both part weight reduction and part consolidation. PP LFTs are the largest segment of this market and are also experiencing the largest growth. Within the PP LFT segment there are three distinct methods of making and using these materials. The classic method is the pultrusion impregnation and/or cross head extrusion method of generating LFT pellets that are subsequently injection molded. For the purpose of this paper this type of material is labeled Granulate Long Fiber Thermoplastics (GLFT). The second main method of producing LFT is In Line Compounding(IC). Within the IC segment there are a number of distinct methods of producing parts. In one method, the IC material is compression molded into parts in a process similar to that used for the manufacture of Glass Mat Thermoplastics and thermoset parts. Typical compression molded parts are made by placing the hot charge in a mold and closing the press quickly to fill or stamp the part. A second method of compression molding uses a plunger tool where the hot charge is placed in a central cavity of the mold and the tool is closed. A plunger then pushes the charge out of the central cavity and into the mold. The parts produced have a different fiber dispersion pattern than what is achieved with the standard compression molding. In a third method the IC material is fed directly into an injection molding machine for either standard Injection molding or Injection compression molding. All three of these IC methods are Direct Long Fiber Thermoplastics (DLFT). The third distinct material within the LFT field is Continuous Long Fiber Thermoplastics (CLFT) and is not discussed here.
In this study, a comparison of parts produced with injection molding, standard compression molding and compression molding with a plunger tool are compared using identical resin formulations, fiber glass type and fiber glass loadings. The fiber glass sizing and the coupling agent loading were varied. In previous studies [1-2] the results of injection molded (IM) PP GLFT parts were studied. The coupling agent melt flow and maleic anhydride loading were found to be important to obtaining optimum properties.
Page 2
180
200
220
240
260
280
300
100
150
200
250
300
0.60.7
0.80.9
1.0
Fle
xu
ral S
tren
gth
(M
Pa)
MFI o
f C
ouplin
g A
gen
t
% MAH in Coupling Agent
0.5% Coupling Agent in Composite
Figure 1: Flexural Strength versus Coupling Agent MFI and MAH Loading In 60% PP GLFT with TufRov® 4575 Fiber Glass
Figure 1 shows the response of flexural strength in 60% glass filled PP GLFT versus the percent of MAH and the MFI of the coupling agent. The percent of MAH is reported as % MAH in the coupling agent. There are a number of trends that can be extracted from Figure 1. First: The maximum flexural strength obtained is almost 280 MPa. This is phenomenally high flex strength for a PP based composite. Second: The higher the MFI of the coupling agent, the higher the flex strength. Third: The higher the MAH loading, the higher the flex strength. The trends break down slightly at low coupling agent loadings but this may be because of an incomplete coverage of the fiber glass with coupling agent.
In the compression molding study [3], the composite properties were not easily affected by changing material variables. The study showed that the DLFT properties for compression molded parts are very stable and predictable even with significant changes in the formulation (Figure 2); even though the fiber glass sizing showed significant differences in adhesion performance in the same parts (Figure 3-4).
Page 3
Figure 2: Coupling Agent Type and Loading versus Tensile Strength in Compression Molded DLFT. No real change in Tensile Strength observed.
Figure 3: Representative curve for TufRov® 4599 with 2% BondyRam 1001
Figure 3: The 4599 single fiber pull out curve follows a classic slip/stick failure mode and is further evidence that the primary mechanism of failure is fiber pull out. Friction is responsible for holding the fiber in the resin. When the load rises to a critical level the fiber slips until the load is alleviated. The cycle then starts again. There is good wetting of the resin onto the fiber surface, but little actual adhesion between the resin and glass.
Me
an
(Te
ns
ile
str
en
gth
(M
Pa
))
0
20
40
60
80
100
cro
ss
-flo
w
flo
w
cro
ss
-flo
w
flo
w
cro
ss
-flo
w
flo
w
cro
ss
-flo
w
flo
w
cro
ss
-flo
w
flo
w
cro
ss
-flo
w
flo
w direction
0.25 0.5 1 1.25 1.5 2 content
Bondyram 1001 coupling
Marlex 120 MFI homo-PP polymer
4575-17mu glass
30
40
glass wt%
Page 4
Figure 4: Representative curve for TufRov® 4575 with 2% BondyRam 1001
Figure 4: The 4575 single fiber pull out test show excellent adhesion between the fiber and polypropylene matrix. The fiber does not slip from the resin but continues to hold until the load reaches a critical level and complete failure occurs. The interface between resin and fiber does not fail and is indicative of a high level of bonding between the fiber glass and resin. The maximum load is also significantly higher and indicative of better tensile and flexural strength in the composite.
Experimental
PPG TufRov 4575, TufRov 4599, TufRov 4588 fiber glass rovings were processed through a Coperion ZSK 40mm twin screw extruder to produce either DLFT compression molded plaques, DLFT plunger tool molded parts, or GLFT pellets for subsequent injection molding. A blended masterbatch containing a heat stabilizer package and carbon black remained constant at 5% (w/w). The amount of coupling agent was varied in the study from 0% to 5% using one coupling agent. The fiber glass type and fiber diameter were varied.
Marlex® HGZ-1200 homopolymer (120MFI)
Polyram BondyRam® 1001 100 MFI Maleic Anhydride grafted PP with 1% Maleic Anhydride (w/w)
A heat stabilizer package
Carbon Black
DLFT Sample Preparation
Both DLFT sample sets were produced by introducing a resin mixture into the twin screw extruder and allowing the resin to melt before directly introducing fiber glass rovings into the extruder at a later point. The heat profile of the twin screw extruder is listed in Table 1.
Page 5
Table 1. Molding Thermal Profile of ZSK 40mm Twin Screw Extruder
The speed of the extruder varied from 60 rpm to 120 rpm while varying the amount of PP resin mixture being fed into the extruder to control the glass content of the samples. Figure 5 illustrates a curve produced to correlate rpm versus lb/hr PP resin to control the percent by weight of fiber glass in the samples.
Figure 5. Correlation of Screw Speed versus Resin Lbs/Hr to control glass content
Table 2. Collection Time of Charge for Screw Speed of Extruder
18
28
38
48
58
50 70 90 110 130
Re
sin
Lb
s/ H
r
Screw Speed (RPM)
Screw Speed vs. Lbs/ Hr
40% Glass Content
30% Glass Content
Molding Profile
Zone 1 170°C
Zone 2 190°C
Zone 3 190°C
Zone 4 220°C
Zone 5 240°C
Collection Times
Screw Speed Glass Content
40% 30%
90 rpm 2min 1min 30sec
60 rpm 2min 40sec 2min
120rpm 1min 30sec 1min 10sec
Page 6
To ensure uniform thickness and consistency in the DLFT produced samples, the amount of extrudate was collected for equal amounts of time depending on the screw speed of the extruder. Table 2 lists the collection times for the different screw speeds.
Figure 6: Plunger Tool
Page 7
Figure 7: Plunger Tool parts without Plaque cavity
Figure 8: A is Injection Molded, B is Plunger Tool and C is Compression Molded. Before and after burn off.
To complete the sample for the compression molded parts the “charge” from the extruder was compression molded. A 40 cm by 40 cm plaque tool was installed in a St. Lawrence 150 ton hydraulic press. The charge was pressed at 450 psi for 40 seconds with a mold temperature of 140°F. The plaques produced were cut using an O-MAX 2626 water jet cutter to make the final test specimens for mechanical analysis.
To complete the sample for the plunger tool molded parts the “charge” from the extruder was rolled into a log and placed in the plunger tool central cavity (Figure 6-7). The St. Lawrence
A B C
Page 8
150 ton hydraulic press was then closed and the hydraulic plunger activated to inject the extrudate into the mold cavities. The cycle time was 72 seconds with a long hold time of 55 seconds after plunger activation. Platen close time was as fast as possible with the press and mold temperature was set at 230 ºF.
GLFT Sample Preparation
GLFT pellets were produced on a cross head pultrusion die attached to the Coperion ZSK 40 mm twin screw extruder. The extruded strands were pelletized to an average length of 12 mm. Glass content was controlled in the cross head die to obtain the target of 40% glass by weight for these studies.
To complete the injection molded parts samples were dried at 120 ⁰C for four hours and the
moisture was measured for each sample prior to molding with maximum allowed moisture of 0.6% (w/w). A Van Dorn 55 injection molding machine was used for molding test specimens. A free flow check valve was installed on the screw tip prior to the initiation of this work. The mold used was standard ISO tensile test specimen mold with large diameter sprues and runners and large radius curves. The gate is a fan gate to the tensile specimens. The mold is a balanced design producing two tensile specimens per shot. Impact and flex specimens were cut from the tensile bars by removing the tab section of the specimen. All test specimens conformed to ISO requirements. The exact profile used for the molding is recorded in Table 3.
Table 3: Van Dorn 55 Injection Molding Profile for GLFT specimens
Molding Profile
Rear Zone 225⁰C
Center Zone 240⁰C
Front Zone 250⁰C
Nozzle 250⁰C
Mold 50⁰C
Shot Size, inch 2.25
Boost Pressure , psi 1000
Injection Speed in/s 1.5
Hold Pressure, psi 925
Screw Speed 90
Back Pressure, psi 50
Unless otherwise stated, all mechanical property testing was performed at 23⁰C and at a
relative humidity of 50%. All testing was done in the PPG Fiber Glass Science and Technology Laboratory in Hoogezand, The Netherlands. Tensile properties were tested according to ISO 527-2 using 10 specimens and a crosshead rate of 5 mm/min (0.2 in/min) with an extensometer gauge length of 50 mm (2 in). Flex properties were tested according to ISO 14125 using 10 specimens and a crosshead rate of 2.0 mm/min (0.8 in/min) and a span of 64 mm (2.56 in). Charpy notched impact properties were tested according to ISO 179-1 with 10 specimens and a Type A notch. Charpy UnNotched impact properties were tested according to ISO 179-1.
Page 9
Results
GLFT Results
The GLFT results follow the trends of previous studies with a nice response to coupling agents loading and obvious differences in fiber performance in the composite (Figure 9-10). The previous trend of 4575 offering the best overall performance is confirmed. As more coupling agent is added the differences in fiber glass performance is reduced with 4588 offering equivalent performance at high loadings of coupling agent. 4575 still offers the best performance in properties and cost as it requires the least amount of the more expensive coupling agent. The 0% loading data does indicate some of the reasons for this improved performance. The 4575 sizing system includes some functional modalities that work as coupling agents. You also see a slight reduction in strengths with increasing fiber glass diameter in the 4575 product. This reduction is often observed in highly crystalline resin systems. The reduction is slight as would be expected in a semi-crystalline PP resin. Figure 11 is a compilation of Tensile and Flexural Strength properties of GLFT samples with 2% coupling agent. This shows some of the differences in fiber glass sizing performance.
Figure 9: GLFT Flexural Properties
Page 10
Figure 10: GLFT Tensile Properties
Figure 11: Flexural and Tensile Strength in GLFT
Page 11
DLFT Compression Molded Results
The DLFT results mimic those obtained in the previous study [3]. There is very little change in performance with either change in fiber glass or the amount of coupling agent added. Statistically the data is equivalent with the possible exception of some differences with no coupling agent added (Figure 12-13). In the case of 0% coupling agent 4588 does poorly and 4599 shows very high impact properties. Because we know the adhesion is not optimal with 4599 the mechanism of failure for the impact is probably changed to fiber pull out for this specific 0% coupling agent sample (Figure 14).
Figure 12: DLFT Flexural Strength
Figures 15 and 16 show a compilation of properties for Tensile and Flexural strengths with a common 2% loading of coupling agent. There is a visual trend of different properties with the different fiber glass samples used. However, the trend is deceiving as there are no statistical differences between these fiber glass reinforcements in the composite.
Flexural Strength for various Fiber Glass Rovings40% PP LFT DLFT Compression Molded
Flow Direction Only
% Coupling Agent in 40% PP LFT
Fle
xu
ral
Str
en
gth
MP
a
0
20
40
60
80
100
120
140
160
180
4575
4588
4599
0% 1% 2% 5%
Page 12
Figure 13: Tensile Strength in Flow Direction for DLFT
Tensile Strength in Flow Direction OnlyFor various Fiber Glass Rovings
40% PP LFT DLFT (Compression Molded)
% Coupling Agent in 40% PP LFT
Ten
sil
e S
tren
gth
MP
a
0
20
40
60
80
100
120
4575
4588
4599
0% 1% 2% 5%
Charpy UnNotched Impact in Flow Direction Onlyvarious Fiber Glass Rovings
40% PP LFT DLFT (Compression Molded)
% Coupling Agent in 40% PP LFT
Ch
arp
y U
nN
otc
he
d I
mp
ac
t k
J/m
2
0
10
20
30
40
50
4575
4588
4599
0% 1% 2% 5%
Figure 14: Charpy UnNotched Impact of Fiber Glass Rovings in DLFT
Page 13
Figure 15: Flexural Strength of Fiber Glass rovings in DLFT
Figure 16: Tensile Properties of Fiber Glass Rovings in DLFT
Page 14
DLFT Plunger Tool Results:
The plunger tool LFT parts are made with a process that is similar to both the DLFT compression molding process and the GLFT injection molding process. The material is heated and mixed in the same system as the compression molded DLFT and is injected into the part cavity in the plunger tool similar to a standard injection molding machine. As such, you would expect that this methodology of part making would produce parts that have aspects of both.
The fiber burn offs shown in Figure 8 tell part of the story. The plunger tool is very gentle on the fiber glass with a very low shear process. The glass lengths obtained are longer on average than what you see in the injection molding process and the fiber alignment is very different from GLFT. In the GLFT parts you have a significant skin and core region of the part. The skin region is highly aligned with the flow while the core fiber glass alignment is more perpendicular to the flow. In the plunger tool parts the alignment appears to be more with the flow and the skin/core effect seen in injection molding was not discernible in the parts we produced with our very gentle system. The fiber lengths obtained in the plunger tool are similar to what was obtained in the compression molded parts. There seems to be a slight loss compared to compression molding, but it may not be significant. Comparison of the fiber alignment between plunger tool parts and compression molded parts indicate that they both have a high degree of alignment in the flow direction. In DLFT compression molding there are very different properties in the flow and the cross flow direction. This is why the flow and cross flow properties are always evaluated to give a better representation of the final part properties. No evaluation of the cross flow properties of the plunger tool parts were made as part of this initial study. If these fiber orientations prove correct, then an evaluation of the cross flow properties will need to be done to give a better representation of the actual composite properties.
Figure 17: Flexural Strength in Plunger Tool Molded parts
Page 15
Figure 18: Charpy UnNotched Impact in Plunger Tool Parts
Figure 19: Tensile Strength in Plunger Tool Parts
Charpy UnNotched Impact on Various Fiber Glass Rovings40% PP LFT Plunger Tool
% Coupling Agent in 40% PP LFT
Ch
arp
y U
nN
otc
he
d I
mp
ac
t k
J/m
2
0
10
20
30
40
50
60
70
4575
4588
4599
0% 1% 2% 5%
Page 16
Figure 20: Flexural Strength in Plunger Tool parts
Figure 21: Comparison of molding method to Tensile Strength
The plunger tool gives impressive mechanical properties that are comparable to those obtained in GLFT. The materials appear to be more sensitive to changes in the formulation of
TufRov 4575 Flexural Strength in 40% PP LFTComparison of Plunger Tool, Injection Molded
and Compression Molded
% Coupling Agent in 40% PP LFT
Fle
xu
ral
Str
en
gth
MP
a
0
50
100
150
200
2504575-Plunger
4575-Injection
4575-Compress Flow
4575-Compress X Flow
0% 1% 2% 5%
Comparison of Plunger Tool to GLFT Injection Moldingto DLFT Compression Molding Flow Direction
TufRov 4575 Sizing System with Different Fiber Glass Diameters
Processing Method versus Fiber Diameter
Ten
sil
e S
tren
gth
MP
a
40
60
80
100
120
140
160
DLFT Plunger Mold GLFT Injection Mold DLFT Compression Mold
2% CC40% PPLFT
17 19 24 24 2417 17 19 19 33
Fiber Glass Diameter
Page 17
the composite, as is seen in GLFT (Figure 17-19). A significantly higher mechanical performance is observed compared to DLFT compression molding (Figure 20-22). The higher fiber alignment in the flow direction is probably contributing to these better mechanical properties. The plunger tool properties are not as sensitive to changes in fiber glass sizing systems as observed in the GLFT data. The overall tensile and flexural properties are higher than for the compression molded samples. The Impact properties may indicate some degree of higher sensitivity to changes in coupling agent loading and fiber glass sizing, but the differences are minor and are not significant. Figures 20-22 show the comparison of properties of a set formulation with 2% coupling agent and with the 4575 fiber glass sizing. The variables evaluated are fiber glass diameter and the molding process for making the parts. The tensile properties in Figure 21 show that the plunger tool gives statistically equivalent performance to the injection molding. The lower flow tensile properties of the compression molded samples are telling. Figure 22 shows the impact performance comparison. The plunger tool impact properties are in between the injection molded and compression molded values. The injection molded Charpy UnNotched impact properties are statistically different from the compression molded parts. The flexural properties in Figure 23 indicate a higher value for the injection molded samples, but this may not be significant. The plunger tool performance is actually very good in tensile, flexural and impact performance.
The evaluation of fiber glass diameter to properties throughout this study indicate that there is little loss in properties with the use of the larger diameter fiber in a polypropylene composite.
A serious study of the plunger tool cross flow properties will need to be undertaken to determine if there is a similar flow/cross flow disparity in properties as is seen in the DLFT compression molded parts. The plunger tool is more expensive to manufacture than a compression molding tool, but the improved performance offered may make it the method of choice for certain LFT part production. In our study we did not optimize cycle time because of the constraints of our press. The long cycle time in this study is probably not indicative what can be obtained with a high speed press.
As a disclaimer: In this study all samples were targeted for a nominal 40% by weight glass content and the compression molded and plunger tool parts were very close to target. In the case of the GLFT parts the glass content on some samples was higher than expected. This will tend to favor the GLFT properties; giving somewhat artificially high numbers, when compared to the other 40% by weight samples. Table 4 shows the average glass content and the range in the samples produced for all three processes. The average values are well within normal variation with good standard deviations however, for the GLFT there were a couple of samples with high weight percent glass loadings that could skew some data. One must be careful in assuming that the results here will be completely indicative of what will be obtained in a production part where the glass content is better controlled. The purpose of this study was to show the potential for greatly improved mechanical properties with the plunger tool compared to standard compression molded DLFT. The GLFT data is included for reference only and was not the focus of the study.
Page 18
Table 4: Glass content range in Data sets
GLFT Injection Plunger Tool DLFT Compression
Average Glass Content (% by weight)
42.89 39.37 38.90
Standard Deviation 1.46 1.55 1.81
Range High 46.56 42.24 43.43
Range Low 40.90 36.66 37.25
Comparison of Plunger Tool to GLFT Injection Moldingto DLFT Compression Molding Flow Direction
TufRov 4575 Sizing System with Different Fiber Glass Diameters
Processing Method versus Fiber Diameter
Ch
arp
y U
nN
otc
he
d I
mp
ac
t k
J/m
2
20
30
40
50
60
70
80
2% CC40% PP LFT
GLFT Injection Mold DLFT Compression Mold
17 19 24 17 19 24 17 19 24 33
Plunger Mold
Figure 22: Comparison of molding method to Charpy UnNotched Impact
Summary and Next Steps
The comparison of plunger tool molded, compression molded and Injection molded LFT parts all using the same formulations has allowed direct comparison of these three distinctly different methods of production. The GLFT parts are easy to mold and give high mechanical properties that are easy to predict. The DLFT parts are easy to mold with fast cycle times and give parts that are very stable in properties with little response to changes in formulation. The plunger tool molded parts are also easy to mold with high mechanical properties in the flow direction and are responsive to changes in the material formulation. It is yet to be seen if the properties observed in the flow direction of the molded parts will translate into higher or lower mechanical properties in the cross flow direction. It is important to know that the plunger tool parts are sensitive to formulation changes and that this can allow optimization of mechanical
Fiber Glass Diameter
Page 19
properties as is seen in GLFT [1-2]. This is significantly different than what has been observed in the compression molded parts [3]. The lack of sensitivity to fiber glass sizing changes in this study with the plunger tool may be indicative of the lower shear process observed and the fiber orientation in the plunger tool. Further work will need to be done to determine how to optimize
the fiber glass performance in the plunger tool process.
Figure 23: Comparison of molding method to Charpy UnNotched Impact
Acknowledgements
Many thanks to Rhonda Hopper and Frank Smith for running the numerous trials needed to generate this study. Also thank you to Piet Leegstra for his work in coordinating the testing.
Bibliography
1. Bowland, C.D., “A formulation study of Long Fiber Thermoplastic Polypropylene (Part One: The effects of coupling agent, glass content and resin properties on the mechanical properties,” ACCE Conference Proceedings (2008).
2. Bowland, C.D., “A formulation study of Long Fiber Thermoplastic Polypropylene (Part Two: The Effects of Coupling Agent Type and Properties,” ACCE Conference Proceedings (2009).
3. Bowland, C.D., Busche, B., Woude, J. vd. “A formulation study of Long Fiber Thermoplastic Polypropylene (Part 3: Mechanical Properties of PP DLFT Composites”, ACCE Conference Proceedings (2011).
4. Allred, R.E., and Wesson, S.P. “Surface Characterization of Sized and Desized Toray M40J Carbon Fibers,” SAMPE Conference Proceedings, (2002).
5. Thomason, J.L., “The influence of fibre length and concentration on the properties of glass fiber
Comparison of Plunger Tool DLFT to GLFT Injection Moldingto DLFT Compression Molding Flow Direction
TufRov 4575 Sizing Sytem with differert Fiber Glass Diameters
Processing Method versus Fiber Diameter
Fle
xu
ral
Str
en
gth
MP
a
80
100
120
140
160
180
200
220
DLFT Plunger Mold GLFT Injection Mold DLFT Compression Mold
2% CC40% PPLFT
17 17 17 19 19 1924 24 24 33
Fiber Glass Diameter
Page 20
reinforced polypropylene: 5. Injection moulded long and short fibre PP,” Composites: Part A 33 (2002) pp. 1641-1652.
6. Lee, J.S and Lee, J.W., “Melt Impregnation Behavior and Mechanical Properties of Long Fiber Thermoplastics Composites through Pultrusion Process,” ANTEC Conference Proceedings, (1996) pp. 2536-2540.
7. Azari, A.D., ”The Influence of the Pultrusion Line Speed on the Mechanical Properties of the Thermoplastic Rovings Impregnated Long Fiber”, ANTEC Conference Proceedings, (1996) pp. 2526-2530.
8. Thomason, J.L., “The influence of fibre length and concentration on the properties of glass fibre reinforced polypropylene: 7: Interface strength and fibre strain in injection moulded long fibre PP at high fibre content,” Composites: Part A 38 (2007) pp. 210-216.
9. Kumar, K.S., Naresh, B., Anup, K.G., ”Mechanical Properties of Injection Molded Long Fiber Polypropylene Composites, Part 2: Impact and Fracture Toughness”, Polymer Composites (2008) pp.525-533.
10. Hong, C.H., Lee, Y.B., Bae, J.W., Jho, J.Y., Nam, B.U., Hwang, T.W., “Molecular Weight Effect of Compatibilizer on Mechanical Properties of Polypropylene/Clay Nanocomposites”, J. Ind. Eng. Chem. Vol. 11, No.2, (2005) pp. 293-296.
11. Cantwell, W.J., Tato, W., Kausch, H.H., “The Influence of a Fiber-Matrix Coupling Agent on the Properties of a Glass Fiber/Polypropylene GMT”, J Thermoplastic Composites Materials, Vol.5, Oct 1992, pp.304-317.
12. Bailey, R.S., Lehr, W., Moore, D.R., Robinson, I.M., Rutter, P.M., “Long Fibre Reinforced Thermoplastics for Injection Moulding: The Relationship between Impregnation and Properties”, Dev. Sci. Technol. Compos. Matter, Eur. Conf. Compos. Mater., 4th (1990), pp.1037-42
13. Bailey, R.S., Moore, D.R., Robinson, I.M., Rutter, P.M., “The effect of impregnation on the microstructure of long fiber-reinforced thermoplastic composites and their consequence on deformation and toughness”, Science and Engineering of Composite Materials (1993), 2(3), pp.171-94.
14. Kim, H.C., “Toughening mechanisms of Long-Fiber-Reinforced Thermoplastics”, Society of Automotive Engineers, [Special Publication] SP (1998), SP-1340(Plastics: Components, Processes, and Technology), pp.167-171.
15. Grove, D.A., Kim, H.C., “Effect of Constituents on the Fatigue Behavior of Long Fiber Reinforced Thermoplastics”, ANTEC Conference Proceedings, (1995), pp. 3003-3007.
16. Thomason, J.L., Schoolenberg, G.E., “An investigation of glass fibre/polypropylene interface strength and its effect on composite properties”. Composites Vol. 25, No 3, (1994), pp.197-203.
17. Thomason, J.L., Groenewoud, W.M., “The influence of fibre length and concentration on the properties of glass fibre reinforced polypropylene: 2. Thermal Properties”, Composites Part A, 27A, (1996), pp. 555-565.
18. Thomason, J.L., Vlug, M.A., “Influence of fibre length and concentration on the properties of glass fibre-reinforced polypropylene: 1: Tensile and flexural modulus”, Composites: Part A 27A, (1996), pp. 477-484.
19. Thomason, J.L., Vlug, M.A., “Influence of fibre length and concentration on the properties of glass fibre-reinforced polypropylene: 4: Impact properties”, Composites: Part A 28A, (1997), pp.277-288.
