Effects of the Chemical Foaming Agents, Injection

Parameters, and Melt-Flow Index on the Microstructure

and Mechanical Properties of Microcellular Injection-
Molded Wood-Fiber/Polypropylene Composites

Andrzej K. Bledzki, Omar Faruk

Institut für Werkstofftechnik, Kunststoff- und Recyclingtechnik, University of Kassel, Mönchebergstraße 3, D-34109
Kassel, Germany

Received 27 May 2004; accepted 14 September 2004

DOI 10.1002/app.21685
Published online in Wiley InterScience (www.interscience.wiley.com).

ABSTRACT: Wood-fiber-reinforced plastic profiles are microscopy) and scanning electron micrographs showed
growing rapidly in nonstructural wood-replacement appli- that an exothermic chemical foaming agent produced the
cations. Most manufacturers are evaluating new alternative best performance with respect to the cell size, diameter, and
foamed composites, which are lighter and more like wood. distance. The polymer melt-flow index and the variation of
Foamed wood composites accept screws and nails better the injection parameters affected the properties and micro-
than their nonfoamed counterparts, and they have other structure of the microfoamed composites. The density of the
advantages as well. For example, internal pressures created microfoamed hardwood-fiber/PP (with a high melt-flow
by foaming give better surface definition and sharper con- index) composites was reduced by approximately 30% and
tours and corners than nonfoamed profiles have. In this decreased to 0.718 g/cm3 with an exothermic chemical
study, the microfoaming of polypropylene (PP) containing foaming agent. Tensile and flexural tests were performed on
hardwood fiber was performed with an injection-molding the foamed composites to determine the dependence of the
process. The effects of different chemical foaming agents mechanical properties on the density and microvoid content
(endothermic, exothermic, and endothermic/exothermic), of the foamed specimens, and these properties were com-
injection parameters (the mold temperature, front flow pared with those of nonfoamed composites. MAH-PP im-
speed, and filling quantity), and different types of PP (dif- proved the physicomechanical properties up to 80%. With
ferent melt-flow indices) on the density, microvoid content, an increase in the mold temperature (80 –110°C), the surface
physicomechanical properties, surface roughness, and mi- roughness was reduced by nearly 70% for the foamed com-
crocell classification of microfoamed PP/wood-fiber com- posites. © 2005 Wiley Periodicals, Inc. J Appl Polym Sci 97:
posites were studied. A maleic anhydride/polypropylene 1090 –1096, 2005
copolymer (MAH-PP) compatibilizer was used with the in-
tention of improving the mechanical properties of micro- Key words: composites; fibers; foams; injection molding;
foamed composites. The microcell classification (from light mechanical properties; poly(propylene) (PP)

INTRODUCTION mer/wood-fiber composites through a batch foaming

process3–7 and a continuous extrusion process8 –12 has
Many new technologies are now being introduced and
been investigated. Park et al.13 experimented with two
reintroduced for foam injection-molding processing.
system configurations (a tandem extrusion system vs a
Often called microcellular foaming, these new technol-
single extruder system) for wood-fiber/polymer com-
ogies use a number of approaches to achieve fine
posites to demonstrate the system effect on the cell
cellular structures with double-digit weights and cycle morphology and foam properties. Microcellular foam-
time reductions. The keys to these innovative technol- ing in an injection-molding process was also intro-
ogies are computerized process control and good tool duced with natural fiber (flax fiber) and wood fi-
design, including counter pressure, static melt mixing, ber.14 –17
and new chemical foaming agents (CFAs).1,2 A recent report18 reviewed the different processing
Moreover, one of the fastest growing materials for systems used for microcellular wood-fiber-reinforced
plastics is also one of the oldest: wood fiber. The polymer composites. It described the recent develop-
processing of microcellular foamed structures in poly- ments of microcellular natural-fiber-reinforced and
wood-fiber-reinforced polymer composites.
There are several variables to consider when an
Correspondence to: A. K. Bledzki (kutech@uni-kassel.de).
injection-molding machine is operated. Some of these
Journal of Applied Polymer Science, Vol. 97, 1090 –1096 (2005) variables can affect the physical properties of foam
© 2005 Wiley Periodicals, Inc. composites. It is well established that the mold tem-
WOOD-FIBER/POLYPROPYLENE COMPOSITES 1091

perature and cooling time are important variables in burg, Germany). In this study, CFAs were used at a 4
this respect. Many other factors can be adjusted, in- wt % concentration in the composites.
cluding variables such as the front flow speed and
filling quantity, that can affect one or more foam prop-
Processing and foaming
erties.
CFAs enable both commodity and engineering Hardwood fibers with PP were mixed in a Henschel
polymers to be processed more easily and make pos- HM40 KM120 high-speed mixer with and without a
sible improved properties for injection-molding and coupling agent. Hardwood fibers were dried at 80°C
extrusion processes. Compatibility and chemical reac- in an air-circulating oven for 24 h (moisture content
tions with foaming agent byproducts are becoming ⬍ 1%) before mixing. The high-speed mixer was pre-
more important because of improvements in the heated to 180°C, the speed of the rotors was set to 2200
weights of molded parts. Also, dramatic strides are rpm, and the mixing time was 12–15 min. Cold ag-
being made with new CFAs in achieving reproducible glomerated granules were then mixed with different
cell distributions and sizes in microcellular foaming. CFAs. Before the foaming in the injection molding, the
CFAs are substances that decompose at processing mixed granules were dried at 80°C for 24 h. The spec-
temperatures, and so gases such as CO2 and nitrogen imens (200 mm ⫻ 90 mm ⫻ 4 mm) of the hardwood-
are liberated. Solid organic and inorganic substances fiber foamed composites were prepared by an injec-
are used as CFAs. The solid residues react as nucle- tion-molding process at a melting temperature of 150 –
ation centers. This leads to a finer cell structure and 180°C, a mold temperature of 80 –110°C, and an
better solubility of the gas in the polymer melt. There injection pressure of 20 kN/mm2.
are several different types of CFAs, which differ
mostly in the type of gas that is generated and the type
Measurements
of reaction that generates the gas. The reaction that
produces the gas can either absorb energy (endother- The tensile and flexural tests were conducted with a
mic) or release energy (exothermic). Generally, CFAs Zwick UPM 1446 at a test speed of 2 mm/min accord-
are divided by their enthalpy of reaction into two ing to EN ISO 527 and EN ISO 178, respectively, for
groups: exothermic and endothermic foaming agents. the hardwood-fiber/PP composites. All tests were
Nowadays, a combination of exothermic and endo- performed at room temperature (23°C) at a relative
thermic CFAs is also used for foaming. humidity of 50%, and five to eight samples were tested
We investigated the effects of different CFAs, injec- for each treatment. The densities of the nonfoamed
tion parameters, and polymer melt-flow indices on the and microfoamed specimens were measured accord-
physicomechanical properties of resultant hardwood- ing to DIN 53479. Fifteen replicates were conducted
fiber/polypropylene (PP) microfoamed composites for each treatment. The microvoid content was calcu-
and the influence of the compatibilizer on the proper- lated according to ASTM Standard D 2734-70 for
ties. foamed composites. The surface roughness of the
wood-fiber/PP composites was measured according
to ISO 4287/1 with a laser-sensor autofocus measure-
EXPERIMENTAL ment system.
Materials
PP [Stamylan P17M10 (melt-flow index ⫽ 10.5 g/10 Scanning electron microscopy (SEM) and light
microscopy
min) and Stamylan P213MNK 40 (melt-flow index
⫽ 90 g/10 min)] was provided as granules by DSM The morphology of the wood-fiber-reinforced micro-
(Gelsenkirchen, Germany). cellular PP composites and the cell size, shape, and
Hardwood fiber (Lignocel HBS 150-500) with a par- distribution of the microcells in the microfoamed com-
ticle size of 150 –500 ␮m was supplied by J. Retten- posites were investigated with SEM (Vega Tescan)
maier & Söhne GmbH (Holzmuhle, Germany). and light microscopy. Cross sections of the sanded
A commercially available maleic anhydride/ and polished surfaces were studied with light micros-
polypropylene copolymer (MAH-PP; Licomont AR copy. Fractured surfaces of the flexural test samples
504 FG; acid number ⫽ 37– 43 mg of KOH/g) was were observed with SEM after they were coated with
used as a compatibilizer. It was obtained from Clariant gold.
Corp. (Frankfurt, Germany). It accounted for 5% of the
weight percentage of the wood fiber.
Microcell classifications
To obtain foamed wood-fiber-reinforced compos-
ites, we used three types of CFAs (endothermic, exo- This investigation was accomplished with a digital
thermic, and endothermic/exothermic), which were image analysis system with Digitrace image process-
obtained from Clariant Masterbatch GmbH (Ahrens- ing software.
1092 BLEDZKI AND FARUK

The image processing and evaluation took place as

follows:

Picture introduction.
Picture improvement.
Object recognition.
Binary image processing.
Object measuring technique.

For the evaluation of the microcell classifications,

the following measured variables from the geometri-
cal relations of the binary picture were presented as
results.

Microcell diameter
The cavity diameter was indicated with respect to the
current scaling in micrometers. The diameter of the
pores was calculated over the surface of a circle with
same area.

Cell distance
This value described the distribution of the microcells.
The distances of the neighboring cells were measured.
With the help of an interactive interference, it was
possible to correct wrongly reconstructed line distance
measurements.

RESULTS AND DISCUSSION

The microstructure of the foamed composites was in-
fluenced by the CFAs in the injection-molding process.
The foaming area and distribution of the cells are
shown in light micrographs of the hardwood-fiber/PP
microfoamed composites in Figure 1(a– d). The micro-
graphs illustrate that the foamed structure, near the
injecting point, had a three-layer sandwich structure.
It contained a middle layer with distributed cells, and
a compact outer hull was identified. Between the
foaming area and surface layers, there was a transition
zone, in which the microcells rode from the injecting
point to the boundary area. Moreover, the foaming Figure 1 Light micrographs of hardwood-fiber/PP micro-
area (middle layer) of the composites with an endo- foamed composites: (a) endothermic PP10, (b) endothermic/
exothermic PP10, (c) exothermic PP10 (melt-flow index ⫽ 10
thermic foaming agent was thinner than that with
g/10 min), and (d) exothermic PP90 (melt-flow index ⫽ 90
other CFAs. The effect of the melt-flow index on the g/10 min). The wood-fiber concentration was 30%, and the
microfoamed structure can be observed in Figure magnification was 12.5:1.
1(c,d). For composites with high-melt-flow-index PP
(PP90), the microcells were bigger in the middle area,
whereas in the boundary area, many smaller micro- composites with endothermic foaming agents. Com-
cells existed. This occurred because PP with a high posites with exothermic foaming agents, however,
melt-flow index improved the expansion of the blis- presented the finest cells. Figure 2 shows SEM micro-
ters in comparison with PP with a lower melt-flow graphs of hardwood-fiber/PP (both types) micro-
index (PP10). foamed composites with an exothermic foaming
Our previous experiment19 demonstrated that CFAs agent. Bigger cells were observed for the high-melt-
play an important role in the cell dimensions and flow-index PP composites rather than the regular PP
distributions. The biggest cell sizes were observed in composites. It is also remarkable that, for the high-
WOOD-FIBER/POLYPROPYLENE COMPOSITES 1093

than PP with a lower melt-flow index because the

viscosity was lower in the boundary area than in the
middle area.
From the light micrographs, the cumulative frac-
tions of the cell diameters and the distances of the
hardwood-fiber/PP microfoamed composites were
measured with Digitrace computer software. Figure 3
shows the cumulative fractions of the cell diameters
with all CFAs for the hardwood-fiber/PP micro-
foamed composites with 30% wood fiber [Fig. 3(a)]
and also the effect of the melt-flow index [Fig. 3(b)].
The cumulative fractions of the cell distances of the
microfoamed composites are presented in Figure 4.
The exothermic foaming agent showed better cell dis-
tribution and form than the other CFAs. PP with a
high melt-flow index showed better performance with
respect to the cell distribution and form.

Figure 2 SEM micrographs of hardwood-fiber/PP micro-

foamed composites: (a) exothermic PP10, (b) exothermic
PP90, and (c) exothermic PP90. The wood-fiber concentra-
tion was 30%.

melt-flow-index PP composites, the cells contained

very small microcells in their walls, as shown in Fig-
Figure 3 Cumulative fraction of the cell diameter of hard-
ure 2(c) in a higher magnification micrograph. This wood-fiber/PP microfoamed composites: (a) PP10 with dif-
was due to the lower viscosity of the high-melt-flow- ferent CFAs and (b) regular PP (PP10) and high-melt-flow-
index PP, which favored the expansion of gases more index PP (PP90). The wood-fiber concentration was 30%.
1094 BLEDZKI AND FARUK

flow-index PP, the composites’ density decreased at

most to 0.718 g/cm3.
In our previous work, we showed that the charac-
teristic mechanical values of hardwood-fiber/PP mi-
crofoamed composites did not differ significantly with
all types of CFAs. Table II presents the specific tensile
and flexural strengths of both types of PP/hardwood-
fiber composites. The specific tensile strength de-
creased proportionately in the foamed composites.
The specific tensile strength and flexural strength for
both types of PP/wood-fiber composites were calcu-
lated from the ratio of the tensile or flexural strength
to the density. With the addition of the MAH-PP (5%)
compatibilizer to the foamed composites, the specific
tensile strength and flexural strength increased for the
regular PP/wood-fiber composites (at most 80% for
the specific tensile strength). However, for the high-
melt-flow-index PP/wood-fiber composites, the spe-
cific tensile and flexural strengths did not increase
more than those of the nonfoamed composites. This
was due to not only the higher microvoid concentra-
tion but perhaps also the larger size of the microcells.
Figure 5 presents the influence of the variation of
the filling quantity on the specific flexural strength.
With the filing quantity increasing, the specific flex-
ural strength decreased. This indicated that a suitable
injected mass should be selected for microfoaming the
materials because a large injected mass increased the
density and at the same time reduced the specific
flexural strength. This confirmed that with this pro-
duction process, raw material savings and improve-
ments in the specific mechanical properties could be
Figure 4 Cumulative fraction of the cell distance of hard-
achieved at the same time.
wood-fiber/PP microfoamed composites: (a) PP10 with dif- Figure 6 shows the influence of the front flow speed
ferent CFAs and (b) regular PP (PP10) and high-melt-flow- on the density and microvoid concentration of the
index PP (PP90). The wood-fiber concentration was 30%. composites. As the front flow speed increased in the
tool, the density decreased and the microvoid concen-
tration decreased. This proves that the viscosity of the
The densities of the microfoamed composites (with injected mass decreased under the influence of the
all the CFAs) and the nonfoamed composites are sum- resulting shear stress from the front velocity of the
marized in Table I. The exothermic foaming agent flow.
reduced the maximum density for the hardwood-fiber For microfoamed products, many researchers and
composites. With the addition of MAH-PP, the density manufacturers of foaming agents have reported that
showed a more reducing tendency, which was around microfoaming produces a smooth surface in compar-
30%, and decreased to 0.741 g/cm3. For high-melt- ison with the surfaces of nonfoamed products.20 Our

TABLE I
Density (g/cm3) of Nonmicrofoamed and Microfoamed Hardwood-fiber/PP composites (30% Wood Fiber)
Exothermic
Endothermic Endothermic/exothermic PP with a high
regular PP regular PP Regular PP melt-flow index

Nonfoamed 1.01 1.01 1.01 0.94

Foamed 0.85 0.78 0.76 0.73
Foamed with MAH-PP 0.84 0.77 0.74 0.71
WOOD-FIBER/POLYPROPYLENE COMPOSITES 1095

TABLE II
Specific Tensile and Flexural Strength of Nonmicrofoamed and Microfoamed Hardwood-fiber composites (30% Wood
Fiber and Exothermic Foaming Agent)
Specific flexural strength
Specific tensile strength [MPa/(g/cm3)] [MPa/(g/cm3)]
PP with a high PP with a high
Regular PP melt-flow index regular PP melt-flow index

Nonfoamed 25.58 24.85 46.9 40.4

Foamed 16.71 15.43 43.5 36.09
Foamed with MAH-PP 29.81 18.37 47.5 39.52

previous results19 showed that the surface roughness CONCLUSIONS

of composites could be significantly improved by mi-
Microcellular hardwood-fiber-reinforced PP compos-
crofoaming. The surface roughness of microfoamed
ites were prepared with an injection-molding process.
composites with an endothermic foaming agent was
This study examined the effects of the different types
reduced by at most 70% with respect to that of non-
of CFAs, the polymer melt-flow index, the variation of
foamed composites.
the injection parameters, and the compatibilizer on the
The mold temperature had a great influence on the
microstructure and mechanical properties of the re-
surface roughness of the microfoamed composites.
sultant foamed composites. The following conclusions
Figure 7 shows laser surface photographs of hard-
were made:
wood-fiber/PP microfoamed composites containing
30% wood fiber. The upper and lower peaks (the
1. The microfoam structure of the composites with
number of peaks and the area of the peaks) indicate
an exothermic foaming agent showed the best
the surface roughness of the composites. With an in-
performance with respect to the cell size, diam-
crease in the mold temperature from 80 to 110°C, the
eter, and distance for the microfoamed compos-
surface roughness decreased for the microfoamed
ites.
composites. The roughness maximum and arithmeti-
2. With the addition of a compatibilizer, the physi-
cal roughness mean deviation of the foamed compos-
comechanical properties of the wood-fiber/PP
ites are illustrated in Figure 8. The surface roughness
microfoamed composites improved up to 80%
of the microfoamed composites was significantly re-
(30% hardwood fiber and an exothermic foam-
duced (at most 70%) as the mold temperature in-
ing agent).
creased. With an increasing mold temperature, the
3. The melt-flow index and the variation of the
temperature difference was reduced between the
injection parameters had a great influence on
foamed core and surface of the composites. Thus, the
the properties and structure of the microfoamed
gas expanded with the mass against the smooth mold
composites:
wall. Moreover, PP exhibited a better surface as the
mold temperature increased.
The density was reduced by approximately 30%
and decreased to 0.71 g/cm3 for the hardwood-

Figure 6 Influence of the front flow speed on the density

Figure 5 Influence of the filling quantity on the specific and microvoid concentration. The hardwood-fiber concen-
flexural strength. The hardwood-fiber concentration was tration was 30%, and the exothermic foaming agent was
30%, and the exothermic foaming agent was regular PP. regular PP.
1096 BLEDZKI AND FARUK

fiber microfoamed composites (high-melt-flow-

index PP) with an exothermic foaming agent.
With an increase in the mold temperature from 80
to 110°C, the surface roughness of the foamed
composites decreased by nearly 70%.

Figure 8 Influence of the mold temperature on the surface

roughness (roughness maximum and arithmetical rough-
ness mean deviation) of microfoamed composites. The hard-
wood-fiber concentration was 30%, and the exothermic
foaming agent was regular PP.

An optimum filling quantity could offer raw mate-

rial savings with better specific mechanical prop-
erties at the same time.

References
1. Reedy, M. E. Proceedings of Blowing Agents and Foaming
Processes, Heidelberg, Germany, May 2002.
2. Schut, J. H. Plast Technol 2001.
3. Matuana, L. M.; Park, C. B.; Balatinecz, J. J. Polym Eng Sci 1998,
38, 1862.
4. Matuana, L. M.; Park, C. B.; Balatinecz, J. J. Polym Eng Sci 1997,
37, 1137.
5. Matuana, L. M.; Park, C. B.; Balatinecz, J. J. J Cellul Plast 1996,
32, 449.
6. Matuana, L. M.; Mengeloglu, F. J Vinyl Addit Technol 2001, 7, 67.
7. Matuana, L. M.; Rachtanapun, P.; Selke, S. E. M. J Appl Polym
Sci 2003, 88, 2842.
8. Matuana, L. A.; Mengeloglu, F. Soc Plast Eng Annu Tech Conf
Tech Pap 2001, 3, 2997.
9. Park, C. B.; Rizvi, G. M.; Zhang, H. Presented at the 5th Inter-
national Conference on Wood Fiber–Plastic Composites, Madi-
son, WI, 1999; 105.
10. Matuana, L. M.; Balatinecz, J. J.; Park, C. B. Presented at the 5th
International Conference on Wood Fiber–Plastic Composites,
Madison, WI, 1999; 318.
11. Matuana, L. M.; Mengeloglu, F. J Vinyl Addit Technol 2001, 7, 142.
12. Matuana, L. M.; Li, Q. J Appl Polym Sci 2003, 88, 3139.
13. Zhang, H.; Rizvi, G. M.; Lin, W. S.; Guo, G.; Park, C. B. Soc Plast
Eng Annu Tech Conf Tech Pap 2001, 2, 1746.
14. Bledzki, A. K.; Faruk, O. Soc Plast Eng Annu Tech Conf Tech
Pap 2002, 2, 1897.
15. Bledzki, A. K.; Faruk, O.; Zhang, W. Presented at the 5th Interna-
tional AVK-TV Conference for Reinforced Plastics and Thermoset
Moulding Compounds, Baden-Baden, Germany, 2002; D5–1.
16. Bledzki, A. K.; Faruk, O. Cellul Polym 2002, 21, 417.
17. Bledzki, A. K.; Faruk, O. Presented at the 7th International
Conference on Wood Fiber–Plastic Composites, Madison, WI,
2003.
18. Bledzki, A. K.; Sperber, V. E.; Faruk, O. Rapra Rev Rep 2002,
Figure 7 Influence of the mold temperature on the surface 13(8), 1.
roughness (laser-sensor photographs) of microfoamed com- 19. Bledzki, A. K.; Faruk, O. Compos Part A, submitted.
posites: (a) 80, (b) 100, and (c) 110°C. The hardwood-fiber 20. Holzberg, T.; Luebke, G. Presented at the 7th Technical Confer-
concentration was 30%, and the exothermic foaming agent ence on New Developments in Extrusion, Wuerzburg, Ger-
was regular PP. many, 2002.