Effects of injection-molding conditions on the gloss and color of pigmented polypropylene.

INTRODUCTION

A satisfactory visual impression of the end product is one of the basic quality criteria for polymer injection molding. The assessment of surface appearance is increasingly important, including not only appearance-related features, such as welds, air traps, sink marks, flow marks, etc. [1], but also gloss, color, and surface texture. This is evident in the automotive industry, and it is partly associated with the aim to use pigmented polymers in increasing amounts, in addition to additives and fillers, rather than to perform surface treatments or painting operations in order to meet the quality appearance target.

The temperature, pressure, and melt flow rate are known to strongly affect the bulk and surface structure of injection-molded polymers. In a typical molding cycle, e.g., in Ref. 2, the polymer melt fills the relatively cold mold cavity at the selected filling rate (injection speed). High pressure (holding pressure) is then applied to obtain a detailed replica of the mold topography and volume. This stage lasts until the material at the critical region within the molding, usually close to the injection gate, solidifies (giving the gate-seal time). A residual cooling time is then required until the molding can be ejected from the mold without distortion. During this cooling stage, the melt volume to be used in the next cycle is plasticized at the selected temperature profile with the required back pressure, screw angular speed, and melt decompression.

The purpose of the present work was to evaluate the effect of the process conditions (materials, mold topography, and processing conditions) on the gloss and color of mineral-filled polypropylene containing different amounts of pigment mixture from 0.25 to 1.5 wt% giving an increasing beige intensity. A rectangular mold containing three different surface patterns (coarse, smooth, and fine) was used. In-mold rheology and a gate-seal analysis were employed to select the filling and postfilling processing windows. The effects of filling rate, holding pressure, and mold temperature on the gloss, color, and surface topography were evaluated. A description of appearance characteristics is given below.

Gloss, Color, and Surface Topography

From a physical point of view, the two quantitative descriptors of appearance (gloss and color) are the consequences of complex psychophysical phenomena of visual perception related to a situation in which the light reflected from the surface of an opaque sample is either predominantly in the specular direction (gloss) or diffuse in all directions (color), compare to Refs. 3 and 4.

The four variables that primarily affect the gloss are the surface topography (or texture) [5], the wavelength and angle of the incident light [6], and the refractive index of the material [4, 7], whereas the color depends on the illumination conditions, the observation angle, the optical characteristics of the material, the amount of the colorant present, the surface topography [8], and the gloss [9, 10].

When light is reflected from a polymer surface, the surface topography is particularly important since the refractive indices of polymers in general vary only within a very narrow range [11]. Smooth surfaces reflect light mainly in the specular direction, whereas diffuse reflection dominates in the case of rough surfaces. In the case of injection-molded items, reflection from rough surfaces is of particular interest since mold surfaces of different textures are often used in order to add value to polymeric products. Low-gloss (matte) surfaces are preferred when a high perceived quality impression of the product is desired, whereas high-gloss (shiny) surfaces may be required not only for aesthetic reasons but also to facilitate cleaning of the product. Different surface textures are often embossed into the same molded item, e.g., Ref. 12. Experimental [13, 14] as well as theoretical [5, 7, 13] work has indicated that there is an inverse relationship between roughness and gloss.

Previous studies on polymer injection moldings have shown that the gloss increases with increasing mold temperature [15-17] (although not always [17]) for some rubber-modified thermoplastics. The effect of the mould temperature was typically much higher than that of the melt temperature, the effect of the latter being negligible for acrylonitrile-butadiene-styrene (ABS) [16]. The gloss was reported to increase with increasing flow length for ABS [15] and for virgin and recovered polypropylene [18], although other studies showed the negative correlation between gloss and flow length for some rubber-modified thermoplastics [17]. An increasing holding time leveled off the gloss for ABS after an initial increase [16], whereas an increasing holding/packing pressure typically increased the gloss for the same rubber-modified thermoplastics discussed so far [15-17]. All the studies examined point to that the cooling time has a negligible effect on the gloss [15, 16, 18]. A positive correlation between filling rate and gloss (negative correlation with the injection time) was reported for the rubber-modified thermoplastics in Refs. 15-17, whereas a negative correlation was recently observed for virgin and recovered polypropylene [18]. It should be pointed out that in most of these works reviewed, the topography of the mold surface or that of the moldings was not reported making difficult to judge the general validity of the results.

In industrial practice, specular gloss is often expressed in relation to the reflection from an ideal polished black surface in the specular direction (cf. ASTM D2457 or ASTM D523). The gloss is measured with a glossmeter and the results are expressed in gloss units (GU), which are calculated as the reflectometer reading for the surface concerned calibrated with respect to that of a standardized black glass plate with a known refractive index. Three incidence angles (20[degrees], 60[degrees], and 85[degrees]) are recommended for gloss measurements (ISO 2813) in addition to the 75[degrees] geometry that is standard in the paper industry. The 60[degrees] geometry is extensively used for gloss measurements in the automotive industry.

For color measurements, spectrophotometers are used, where spectral reflectance data are used together with a selected standard illuminant (D65) and observer (10[degrees]) to determine the CIELAB color coordinates (color system defined by the International Commission on Illumination); L* (measure of lightness/darkness), a* (measure of redness/greenness), and b* (measure of yellowness/blueness) [19]. The color measurements can usually be performed in two modes: specular-component-included (SCI) or specular-component-excluded (SCE), which take into account either the total reflectance (specular and diffuse) or primarily the diffuse contributions, respectively [3, 8].

Different techniques are available for the topographical characterization of surfaces depending on the scale range and the resolution of interest. In interior car components, this means a range from one to hundreds of micrometers in height and a resolution from fractions of millimeters up to several millimeters in the lateral dimensions. These methods can be classified mainly as mechanical contact stylus, optical noncontact scanning, and microscopy techniques [20, 21].

Various descriptors can be used to quantitatively characterize the topography of a surface. Traditionally, the so-called P-parameters or R-parameters are obtained from the profile of the surface heights during single (line) or multiple (plane) measurements (ISO 4287), respectively. The rapid development of computing capabilities has facilitated the use of three-dimensional (3D) parameters to characterize the surface topography [21]. The most commonly used 3D-descriptor in the height direction is the root-mean-square roughness [S.sub.q] over an analysis area A,

[S.sub.q] = [square root of ([1/A][[integral].sub.A][z.sup.2](x, y) dx dy)], (1)

where z(x, y) is the surface height measured at the coordinates (x, y) within the area A.

Spatial wavelength analytical techniques can be used to characterize the surface topography in more detail. Some examples of these analyses include autocorrelation functions, power spectral densities, and fractional analysis [22, 23]. The latter is the only method considered here since it provides the information required for the present purpose in a fairly simple manner. In order to perform such an analysis, a height-to-height correlation function or a heights-difference function [C.sub.z]([[lambda].sub.s]) must first be evaluated [22]. [C.sub.z]([[lambda].sub.s]) is defined as the mean square height fluctuation of the surface as a function of the horizontal length scale (or spatial wavelength) [[lambda].sub.s]. For a one-dimensional height profile, this can be expressed as

[C.sub.z]([[lambda].sub.s]) = <(z(x + [[lambda].sub.s]) - z(x))[.sup.2]>. (2)

The height-to-height correlation function [C.sub.z] is a power law function of [[lambda].sub.s] (partly associated with a so-called fractional behavior) followed by a plateau that is related to the roughness value [S.sub.q]. The correlation function is characterized by two correlation lengths that define the maximum length scales below which the fractional behavior is observed. They can be determined from the coordinates of the transition point, or corner wavelength, of [C.sub.z]. The first is a lateral correlation length, [[xi].sub.[parallel]], defined as the wavelength at which [C.sub.z] deviates from the power law behavior. The second, [[xi].sub.[perpendicular to]], is amplitude-related (in the z-direction), and it is calculated from the value at which [C.sub.z] attains a constant value corresponding to [[xi].sub.[perpendicular to].sup.2] for [[lambda].sub.s] > [[xi].sub.[parallel]], see also Refs. 22 and 23.

EXPERIMENTAL

Characterization of the Materials

The selected materials were obtained by mixing a heterophasic polypropylene (PP), POLYfill TS20020UV (Polykemi AB, Sweden), with a mixture of pigments (titanium dioxide, heucodur yellow G9101, iron oxide, and carbon black) giving a beige color with an intensity that depends on the amount of pigments added. The polymer matrix was light-stabilized and it contained 10 wt% of talc and 10 wt% of calcium carbonate, according to the supplier. Other characteristics were a density of 1.04 g/[cm.sup.3] (ISO 1183), a melt mass-flow rate at 210[degrees]C of 20 g/10 min (ISO 1133), a tensile modulus at 23[degrees]C of 2000 MPa (ISO 178), a Charpy-notched impact resistance at 23[degrees]C of 5 mJ/[mm.sup.2] (ISO 179), a yield strength of 30 MPa (ISO 527), and a heat distortion temperature (HDT, 120[degrees]C/h at 1820 kPa) of 67[degrees]C (ISO 75). The total amounts of pigment were 0.25 and 1.5 wt%, and the compounded polypropylenes were denoted as PP-1 and PP-2, respectively. The polypropylene without the pigment was denoted PP-0.

A Perkin-Elmer-DSC7 Power-Compensated Differential Scanning Calorimeter (DSC) was used to measure the melting temperature ([T.sub.m]), the degree of crystallinity ([%.sub.c]) and the crystallization temperature ([T.sub.c]) of PP-1 and PP-2, and of the nonpigmented material. Specimens in the weight range 4-8 mg were obtained by compression molding of the polymer pellets at 170[degrees]C using a load of 300 kN (30 tons) for 30 s (Bucher-Guyer press); nitrogen was used as purge gas at a flow rate of 20 ml/min; the heating rate of the first heating ramp was 10[degrees]C/min from 30 to 200[degrees]C followed by equilibration of the molten material at 200[degrees]C for 2 min and then cooling at 10[degrees]C/min from 200 to 30[degrees]C; the heating rate in the second heating was 10[degrees]C/min from 30 to 200[degrees]C. An indium standard was used to calibrate the thermometer and the enthalpy. The heat of fusion for crystalline PP used to calculate the degree of crystallinity was 209 J/g (8.79 kJ/mol) [24].

Table 1 shows the results of the calorimetric measurements. Each value in Table 1 is the average of three measurements with the corresponding standard deviation. From the results obtained with the second heating ramp, it is clear that the melting point [T.sub.m] was virtually unaffected by the pigment added, whereas there was a slight tendency for the crystallinity to decrease with increasing amount of pigment.

Figure 1 shows the shear viscosity as a function of shear rate for PP-1 (dashed line), PP-2 (solid line), and PP-0 (dotted line) at 240[degrees]C. A single-bore capillary rheometer (Rheoscope 1000; Ceast) was used with two capillary dies having length/diameter ratios of 10 and 40 (diameter = 1 mm). The capillary data were corrected according to the Bagley and Rabinowitsch methods using a power-law model with a power-law index of 0.5. At rates exceeding approximately 300 [s.sup.-1], the viscosities of the three materials were essentially identical. This was also the case for PP-1 and PP-2 at lower shear rates, but the nonpigmented PP-0 exhibited a somewhat lower viscosity in this region.

Mold Cavity and Processing

The mold was a rectangular cavity with width (W) 138 mm and length (L) 78 mm as shown schematically in Fig. 2. The mold thickness was 2.7 mm. The mold contained three different surface patterns, type C (coarse), S (smooth), and F (fine), each covering an area of (W/3) X L. The visibly smooth (mirror-like surface) was in the central region of the mold cavity. The mold had a film-edge gate 123 mm wide, 2 mm long, and 1 mm thick.

[FIGURE 1 OMITTED]

The injection-molding machine was an Engel ES 330/80 equipped with a general-purpose screw of 35 mm diameter. The amplification (intensification) factor of the hydraulic pressure at the nozzle was 10. Other characteristics of the machine include a maximum plastification stroke of 160 mm, a maximum angular screw speed of 400 rpm, a maximum specific injection pressure of 1600 bar ([approximately equal to]160 MPa), and a maximum injection speed of 156 mm/s.

[FIGURE 2 OMITTED]

The melt pressure in the nozzle ([P.sub.MELT]) was measured with a Dynisco DST465HXL pressure transducer and sampled at 50 ms intervals with an AAC-2f data logger from Intab Interface-Teknik AB, Sweden. The melt temperature in the nozzle, the screw position, the hydraulic pressure, the switchover point to the holding stage, the filling time, and the plastification stage parameters (back pressure, angular screw speed, and plastification time) were also monitored. A GWK TT70 unit was used to control the temperature of the heat-transfer oil (Statoil Thermway X) circulating transversally with respect to the melt flow direction inside the two heating channels of each mould half (fixed and moving).

Table 2 shows some of the processing parameters used for the materials. The parameters shown were kept constant in this study and they mainly followed the recommendation from the material supplier. The processing parameters that were varied are commented on later.

The chosen molding strategy was based on a decoupling of the filling stage from the holding and residual cooling stages. The residual cooling stage was defined as the time that elapsed from the end of the holding stage to the mold opening and it was concomitant to the plastification stage.

During the filling stage, in-mold rheology [25, 26] was used to obtain the filling window of PP-1 and PP-2 with regard to the mold cavity used with the processing conditions given in Table 2. Figure 3 shows the influence of the injection speed on the apparent shot resistance to fill (ASRF) of molten PP-1 and PP-2 required to fill 95% of the mold cavity volume (squares). Each ASRF-value was obtained by multiplying the melt peak pressure ([P.sub.MELT]) by the corresponding injection (filling) time obtained at the selected injection speed. The switch over to the holding stage was based on the screw position [27]. In Fig. 3, the open symbols refer to a mold temperature of 30[degrees]C, whereas the closed symbols relate to the higher mold temperature of 60[degrees]C. Figure 3 also shows the shot weight (triangles) as a function of the injection speed; each data point corresponded to the total weight of the polymer in the sprue, the gate, and the 95% filled cavity.

In this case, the same ASRF-curves were obtained for PP-1 and PP-2. The mold temperature had no significant effect on the apparent flow resistance of the materials selected.

Three filling conditions, 5 and 30 mm/s in the region preceding the plateau of the ASRF curves (unstable region), and 120 mm/s (stable region), were selected to give three groups of moldings. These filling conditions gave average injection times of 6.3 s (5 mm/s), 0.9 s (30 mm/s), and 0.3 s (120 mm/s), regardless of the pigment content used or the mold temperature selected.

During the holding stage, the time needed to solidify the gate (gate-seal time) was obtained for each group of moldings in accordance with the gate-seal method [28]. Two postfilling hydraulic pressures of 30 bar ([approximately equal to]3 MPa) and 60 bar ([approximately equal to]6 MPa), corresponding roughly to 300 and 600 bar in terms of specific injection pressure (approximately the melt pressure), were used. Experimental trials with holding pressures higher than 60 bar resulted in mould flashing. The weight of the pigmented polypropylene moldings (filled cavities only) is shown as a function of the holding time using a mold temperature of 30[degrees]C in Fig. 4 and using a mold temperature of 60[degrees]C in Fig. 5. Each point in Fig. 4 and Fig. 5 is the average of 10 measurements with an insignificant standard deviation. The solidification of the gate, corresponding to the leveling of the curves in Fig. 4 and Fig. 5, occurred after 13 s at a mold temperature of 30[degrees]C and after 15 s at a mold temperature of 60[degrees]C, regardless of the filling rate or the polypropylene grade selected.

[FIGURE 3 OMITTED]

The total cooling time ([t.sub.C]) was defined as the sum of half the fill time ([t.sub.1/2F]), the gate-seal time ([t.sub.S]) and the time required for the part to be ejected without distortion and, simultaneously, for the plastification stage to be completed (residual cooling time, [t.sub.RC]). An approximate solution to the transient conductive heat-transfer problem for a regular rectangular geometry, Eq. 3 [29], was used as an estimate of the total cooling time ([t.sub.C]):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (3)

where t is the thickness (2.7 mm), [alpha] is the effective thermal diffusivity (0.066 [mm.sup.2]/s at a mold temperature of 30[degrees]C and 0.062 [mm.sup.2]/s at a mold temperature of 60[degrees]C) [30], [T.sub.EJE] is the ejection temperature (80[degrees]C), and [T.sub.MELT] is the melt temperature (240[degrees]C).

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

The residence time of the molten polymer in the barrel was kept constant for each group of moldings manufactured at the different injection speeds by selecting an appropriate delay time (elapsing concurrently with the residual cooling time, [t.sub.RC]) of the plastification stage and by keeping the material cushion constant (Table 2).

Analysis Methods

The molded specimens were conditioned at 23[degrees]C and at 50% relative humidity (ISO 291) for 24 hours prior to further analysis at 21 [+ or -] 2[degrees]C described in the following.

The gloss at 60[degrees]C (ASTM D523) was measured using a portable BYK-Gardner micro-TRI-gloss-[mu] Glossmeter calibrated with a standard black glass. The repeatability of the glossmeter was 0.1 GU. An average of five gloss measurements was used in each case. The analysis area was (9 X 18) [mm.sup.2]. The glossmeter was placed longitudinally with regard to the mainstream flow direction and centered on each of the three different surfaces in Fig. 2.

A portable Datacolor Microflash Spectrophotometer was used to obtain the color coordinates (L*, a*, and b*), based on spectral data at 10-nm intervals from 400 to 700 nm (visible spectrum). The analysis area had a diameter of 18 mm. The instrument repeatability was 0.001 CIELAB units. The color measurements were performed in the SCE mode to minimize the effect of the specular reflectance. The moldings examined were those used for the gloss measurements.

An UBM Optac 2000 Optical Profilometer was used to characterize the two rough (type F and C) surfaces of the mold and of the moldings. The instrument was equipped with an auto-focus sensor UBC14 which allowed a maximum measurable height of [+ or -]500 [micro]m at a resolution of 0.12 [micro]m. An analysis area of (16 X 16) [mm.sup.2] was scanned at a speed of 0.5 mm/s. A total of 81 profiles at a distance of 0.2 mm from each other were taken. The sampling interval within each profile was 0.01 mm.

The topography of the smooth surfaces of the moldings was evaluated with a Talystep Mechanical Profilometer equipped with a spherical diamond tip of radius 15 [micro]m and allowing profile root-mean-square roughness measurements in the range 0.04 nm to 4.0 [micro]m. The stylus load was very low (1 mg) in order to minimize possible damage to the polymer surface [31]. Five profiles of 1-mm length were measured. A Wyko RST Plus Interferometer in the phase-shifting (PSI) mode was employed to characterize the topography of the smooth surface of the mold. The PSI mode uses white light filtered to produce red light at a wavelength of about 632 nm, and it uses the phase shift of the interference fringes to measure the surface topography. It is characterized by a vertical measuring range of 160 nm and a vertical resolution of 0.3 nm for a single measurement and less than 0.1 nm for multiple measurements. The lateral range and resolution are limited by the chosen magnification objective and the array size, and they vary from 59 X 45 [[micro]m.sup.2] up to 4.6 X 3.5 [mm.sup.2], and from 12.5 [micro]m/pixel down to the diffraction limit ([approximately equal to]0.5 [micro]m/pixel), respectively. The magnification objective used was 5.3 units. The total sample area measured was 0.9 X 1.2 [mm.sup.2] with a total of 238 profiles (separated from each other by 3.7 [micro]m) and a sampling interval within each profile of 3.2 [micro]m.

For all topographical measurements, the profiling direction was perpendicular to the flow direction.

For the mold, [S.sub.q] was 33.4 [micro]m for the surface type C, 6.1 [micro]m for the F-surface (by optical profilometry) and 0.011 [micro]m for the S-surface (by interferometry). The lateral correlation lengths ([[xi].sub.[parallel]]) were 400 and 78 [micro]m for the C and F surfaces, respectively.

RESULTS AND DISCUSSION

Gloss

The influence of the injection speed (5, 30, and 120 mm/s), the mold temperature (30 and 60[degrees]C), and the holding pressure (30 and 60 bar, hydraulic pressures) on the gloss of the different PP-2 surfaces is shown in Fig. 6 (S), Fig. 7 (F), and Fig. 8 (C). The total amount of pigment (0.25 and 1.5 wt%) added to the polymer did not influence the gloss, i.e., the gloss values of PP-1 and their dependence of the processing conditions were very similar to those of PP-2.

The effect of the processing conditions on the gloss level was most evident on the smooth (type S) surfaces and it was quite drastic. Increases in injection speed, mold temperature and holding pressure led to gloss increases of 185, 75, and 20% for PP-1 and of 255, 80, and 16% for PP-2 (Fig. 6). The highest gloss value (79.8 GU for PP-1 and PP-2) was obtained with the highest injection speed (120 mm/s), a mold temperature of 60[degrees]C and a holding pressure of 60 bar, whereas the lowest gloss values (24.7 GU for PP-1 and 20.3 GU for PP-2) were obtained at the lowest injection speed (5 mm/s), a mold temperature of 30[degrees]C and a holding pressure of 30 bar. In general, the injection speed had the greatest effect on the gloss level.

[FIGURE 6 OMITTED]

For the rough surfaces, which exhibited much lower gloss levels, a decrease in gloss of 27% for PP-1 surface F (30% for PP-2 in Fig. 7) and of 23% for PP-1 surface C (32% for PP-2 in Fig. 8) were noted when the injection speed was increased from 5 to 120 mm/s. The gloss leveled out at high injection speeds, high mold temperatures, and high holding pressures, to 3.6 GU for the PP-1 molding F (3.5 GU for the PP-2 molding F) and to 1.9 GU for the PP-1 molding C (1.7 GU for the PP-2 molding C). An increase in mold temperature led to a decrease in the gloss of 8.2% for PP-1 molding F (14% for PP-2 in Fig. 7) and of 7.7% for PP-1 molding C (12% for PP-2 in Fig. 7) at an injection speed of 5 mm/s. The influence of holding pressure was significant only for the molding C. A decrease in the gloss as high as 16% for PP-1 (20% for PP-2 in Fig. 8) was obtained with increasing holding pressure at an injection speed of 5 mm/s. Another characteristic of the type C moldings was an increase in gloss at an injection speed of 5 mm/s (to a large extent) and 30 mm/s (to a less extent) and a holding pressure of 30 bar when the mold temperature was increased to 60[degrees]C (Fig. 8).

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

The gloss is related to the surface topography, which is in turn a close replication of the mold surface. Table 3 (smooth surfaces), Table 4 (fine surfaces), and Table 5 (coarse surfaces) show the effect of filling rate on the topography descriptors ([S.sub.q] and [[xi].sub.[parallel]]) of selected PP-2 moldings. For comparison, the surface descriptors of the mold are also included. The fine surfaces exhibited in all cases a Gaussian distribution of surface heights, whereas the coarse surfaces were characterized by a more complex distribution [23].

Provided the different surfaces were treated separately, correlations were found between the two surface topography parameters and the gloss when the rate of injection, the mold temperature and the holding pressure were changed. For smooth PP-2 surfaces, the surface roughness expressed in terms of [S.sub.q] decreased drastically when the injection speed increased (Table 3), whereas, for the fine regions of the PP-2 moldings, the [S.sub.q]-values increased when the injection speed increased (Table 4). These changes in the roughness descriptors corresponded to an increase (Fig. 6) and a decrease in gloss (Fig. 7). The magnitude of the surface descriptors obtained for the rough surfaces (fine and coarse) differed, as expected, from each other and from those obtained for the smooth surface. For both the fine and coarse textures, the [S.sub.q]-values were slightly smaller than the corresponding values for the steel mold, indicating a nonperfect reproduction of the mold surface. The polymer melt was apparently not able to reproduce the fine details of the smooth surface of the steel mold either. The lateral correlation lengths of the fine polymer surfaces were larger than those of the mold. In the case of the coarse surface, the [[xi].sub.[parallel]]-values were much smaller than that of the mold.

The results indicate that the rheological properties of the melt have a strong influence on the gloss value. It may be argued that the gloss of the specimens is associated with a replication of the mold surface. A better replication is achieved with a lower melt viscosity at higher shear rates (higher injection speeds or lower injection times) and higher melt and mold temperatures. This gives a higher gloss in smooth surface regions and a lower gloss in textured regions F and C. This is in agreement with the results reported by Edwards and Choudhury [17] for "polished" surfaces, as well as with those reported by Pisciotti et al. [18] for visibly rough surfaces. An increase in the holding pressure has a similar effect, i.e., it improves the ability to replicate the mold surface.

Color

The amount of pigment, the surface texture (S, F, or C) and the processing conditions (injection speed, mold temperature, and holding pressure) had a significant effect on the lightness L*, Fig. 9 (S) and Fig. 10 (F), and on b*, Fig. 11 (S) and Fig. 12 (F). The results relating to the C surfaces were generally very similar to those of the F surfaces. A constant average a*-value of 2.6 units was obtained for PP-1 (3.3 units for PP-2), regardless of the surface pattern or the processing history.

The lightness (L*) of the smooth (S) PP-1 and PP-2 surfaces decreased with increasing injection speed, e.g., in Fig. 9 the decrease in L* for PP-1 from 65.4 to 64.4 units (mold temperature of 30[degrees]C, holding pressure of 30 bar) when the injection speed was increased from 5 to 120 mm/s. L* also decreased when the holding pressure was increased, e.g., in Fig. 9 a decrease in L* for PP-2 from 65.1 to 64.8 units (injection speed of 5 mm/s, mold temperature of 60[degrees]C) and with increasing mold temperature, e.g., in Fig. 9 a decrease in L* for PP-1 from 65.4 to 65.1 units (injection speed of 5 mm/s, holding pressure of 30 bar).

For the rough surfaces F (Fig. 10), a modest increase in lightness was instead observed with increasing injection speed, e.g., in Fig. 10 an increase in L* from 66.6 to 66.8 units (mold temperature of 30[degrees]C, holding pressure of 30 bar). The effect of injection speed on L* was somewhat more pronounced for PP-2 than for PP-1, e.g., in Fig. 10 an increase in L* for PP-2 from 63.9 to 64.5 units (mold temperature of 30[degrees]C, holding pressure of 30 bar). For both the PP-grades, the mold temperature and the holding pressure had a smaller effect than the injection speed on L*. The measured L*-values for the C-surfaces were generally 0.5 units higher than those of the corresponding F-surfaces. The change in lightness due to a change in processing conditions can also be traced to the surface characteristics of the moldings. Somewhat simplified, it can be argued that the fraction of diffusely reflected light increases if the surface roughness increases, and this corresponds to a lower lightness [9]. Again, this emphasizes the importance of achieving adequate surface replication when aiming for a desired appearance.

As expected, an increasing pigment content led to a significant decrease in the lightness, regardless of the surface texture, e.g., in Fig. 9 (S-type) a decrease in L* from 65.4 units (PP-1) to 63.0 units (PP-2) (mold temperature of 30[degrees]C, holding pressure at 30 bar) or in Fig. 10 (F-type) a decrease in L* from 66.6 (PP-1) to 63.9 units (PP-2) (mold temperature of 30[degrees]C, holding pressure at 30 bar).

The b*-coordinate for the PP-1 and PP-2 surfaces of type S increased with increasing injection speed, mold temperature and holding pressure (Fig. 11). For example, Fig. 11 shows an increase in b* for PP-1 from 11.1 to 11.5 units with increasing injection speed at a constant mold temperature (30[degrees]C) and holding pressure (30 bar). In contrast, the F-surfaces exhibited a decrease in b* with increasing injection speed, e.g., in Fig. 12 a decrease in b* for PP-1 from 10.8 to 10.6 units (for PP-2 from 11.2 to 10.9 units), regardless of the mold temperature or holding pressure. As expected, the b*-values increased with increasing amount of pigment regardless of the type of surface (smooth or rough).

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

As previously reported for ABS by Dawkins et al. [15] positive correlations can be established between the gloss values (Figs. 6-8) and the color coordinate b* (Figs. 11 and 12), whereas L* (Figs. 9 and 10) is negatively correlated with both b* (Figs. 11 and 12) and gloss (Figs. 6 to 8). This shows that the gloss (the surface texture) has a significant effect on the color and, in particular, that all the factors that contribute to an increase in gloss have concomitant effects of increasing the color coordinate b* and decreasing the lightness L*. The former is of course only valid for the materials studied here.

[FIGURE 11 OMITTED]

[FIGURE 12 OMITTED]

CONCLUSIONS

The process conditions had a strong influence on the gloss development and on the color of pigmented PP. The rate of filling was the most significant parameter affecting the gloss compared with the influence of the mold temperature and holding pressure. A better replication of the mold texture can be achieved at a lower melt viscosity at higher shear rates (higher injection speeds or shorter injection times) and higher mold temperatures. This gives a higher gloss in smooth regions and a lower gloss in textured regions. An increase in the holding pressure had an effect similar to but smaller than that of the filling rate and processing temperatures. The gloss (or rather the surface texture) had a significant effect on the color: all the factors that contributed to an increase in gloss showed concomitant effects of increasing the color coordinate b* and of decreasing the lightness L*.

ACKNOWLEDGMENTS

We thank Polykemi AB for supplying the materials. The Department of Interior and Climate Engineering at Volvo Car Corporation and Lear Corporation AB in Fargelanda are gratefully acknowledged for technical support. Prof. L. Mattsson at the Department of Production Engineering (Royal Institute of Technology, Stockholm) is gratefully acknowledged for discussions and Talystep measurements. We thank Dr. J. A. Bristow for the linguistic revision of the paper.

REFERENCES

1. M.F. Lacrampe and J. Pabiot, J. Injection Molding Technol., 4, 167 (2000).

2. R.G. Speight, J. Injection Molding Technol., 1, 25 (1997).

3. R.S. Hunter and R.W. Harold. The Measurement of Appearance, John Wiley & Sons, New York (1987).

4. P.M. Morse, "Characterization and Physical Relationships," in Pigment Handbook, Vol. III, T.C. Patton, ed., John Wiley & Sons, New York, 341 (1973).

5. D.J. Withehouse, D.K. Bowen, V.C. Venkatesh, P. Lonardo, and C.A. Brown, Cirp Annals., 43, 541 (1994).

6. B. Donald and R. Mathew, SPE Antec Tech. Paper, 34, 18 (1988).

7. J.M. Bennett and L. Mattsson, Introduction to Surface Roughness and Scattering, Optical Society of America, Washington, DC (1989).

8. K. Huff, Visual Assessment and Practical Colorimetry in the Plastic Industry, Bayer AG, Leverkusen. Germany (1994).

9. E.N. Dalal and K.M. Natale-Hoffman, Color Research and Application, 24, 369 (1999).

10. I. Arino, U. Kleist, and M. Rigdahl, Polym. Eng. Sci., 45(5), 733 (2005).

11. J.C. Seferis, in Polymer Handbook, 3rd ed., Vol. VI, J. Brandrup and E.H. Immergut, eds., John Wiley & Sons, New York, 451 (1989).

12. D. Schauf, "Reproducing Textures from the Cavity Surface to the Surface of the Thermoplastic Moulding," Application Technology Information ATI-584e, Bayer MaterialScience AG Leverkusen, Germany, KU-Martketing Technische Redaktion (1988).

13. L. Wang, T. Huang, M.R. Kamal, A.D. Rey, and J. Teh, Polym. Eng. Sci., 40, 747 (2000).

14. I. Arino, U. Kleist, L. Mattsson, and M. Rigdahl, Polym. Eng. Sci., 45(10), 1343 (2005).

15. E. Dawkins, P. Engelmann, and K. Horton, J. Injection Molding Technol., 2, 1 (1998).

16. K. Koppi, J.M. Ceraso, J.A. Cleven, and B.A. Salamon, SPE Antec Tech. Paper, 48, 184 (2002).

17. S.A. Edwards and N.R. Choudhury. Polym. Eng. Sci., 44, 96 (2004).

18. F. Pisciotti, A. Boldizar, and M. Rigdahl, Int. Polym. Proc., 20(3), (2005).

19. CIE Publication No. 15.2, Colorimetry, CIE, Vienna (1986).

20. K.J. Stout, Development of Methods for the Surface Characterization of Roughness in Three Dimensions, Penton Press, London (2000).

21. T.R. Thomas, Rough Surfaces, Imperial College Press, London (1999).

22. M. Kluppel and G. Heinrich, Rubb. Chem. Technol., 73, 578 (2000).

23. I. Arino, U. Kleist, G.G. Barros, P.-A. Johansson and M. Rigdahl, Polym. Eng. Sci., 44(9), 1615 (2004).

24. R.P. Quirk and M.A.A. Alsamarraie, in Polymer Handbook, 3rd ed., Vol. V, J. Brandrup and E.H. Immergut, eds., John Wiley & Sons, New York, 27 (1989).

25. J. Sloan, Injection Molding Mag., 5(10), 118 (1997).

26. F. Pisciotti, A. Boldizar, and M. Rigdahl, Int. Polym. Proc., 20(3), (2005).

27. R.A. Malloy, S.J. Chen, and S.A. Orroth, SPE Antec Tech. Paper, 33, 225 (1987).

28. G. Menges, Plastverarbeiter. 29, 369 (1978).

29. D.V. Rosato, D.V. Rosato, and M.G. Rosato, Injection Molding Handbook, Kluwer Academic Publishers, Boston, 321 (2000).

30. R.A. Malloy, Plastic Part Design for Injection Molding, Hanser Publishers, New York, 86 (1994).

31. L. Mattsson and P. Wagberg, Precision Eng., 15(3), 141 (1993).

Francesco Pisciotti, Antal Boldizar, Mikael Rigdahl

Department of Materials Science and Engineering, Chalmers University of Technology, SE-412 96 Goteborg, Sweden

Ingrid Arino

Department of Materials Science and Engineering, Chalmers University of Technology, SE-412 96 Goteborg, Sweden

Department of Interior and Climate Engineering, Volvo Car Corporation, SE-405 31 Goteborg, Sweden

Correspondence to: A. Boldizar; e-mail: [e-mail]

Contract grant sponsor: MARCHAL, Industrial Graduate School in Materials Science, Chalmers University of Technology.

TABLE 1. The calorimetric analysis of PP-0, PP-1, and PP-2.

                    PP-0                            PP-1

                    [T.sub.m] = 157.5 [+ or -] 1.3  [T.sub.m] = 155.7
                                                      [+ or -] 0.9
First heating (a)   [%.sub.c] = 27.8 [+ or -] 0.7   [%.sub.c] = 27.3
                                                      [+ or -] 0.5
Cooling (a)         [T.sub.c] = 127.1 [+ or -] 0.3  [T.sub.c] = 127.4
                                                      [+ or -] 0.2
                    [T.sub.m] = 159.8 [+ or -] 0.1  [T.sub.m] = 159.5
                                                      [+ or -] 0.1
Second heating (a)  [%.sub.c] = 30.4 [+ or -] 0.7   [%.sub.c] = 28.4
                                                      [+ or -] 0.8

                    PP-2

                    [T.sub.m] = 154.3 [+ or -] 0.8
First heating (a)   [%.sub.c] = 26.7 [+ or -] 0.7
Cooling (a)         [T.sub.c] = 127.3 [+ or -] 0.1
                    [T.sub.m] = 159.6 [+ or -] 0.1
Second heating (a)  [%.sub.c] = 27.8 [+ or -] 0.5

(a) The melting and crystallization temperatures are in degrees Celsius
([degrees]C).

TABLE 2. The processing parameters selected for PP-1 and PP-2.

Cylinder
temperature profile  [T.sub.MOLD]  Plastification parameters
([degrees]C)         ([degrees]C)  Hydraulic back pressure

190-215-225-235-     30.60 (b)     15 bar
  240 (a)                            ([approximately equal to]1.5 MPa)

Cylinder
temperature profile                Plastification parameters
([degrees]C)         Angular screw speed  Decompression  Cushion

190-215-225-235-     100 rpm              5 mm           5 mm
  240 (a)

(a) The temperature profile started from the feeding zone of the
cylinder increasing up to 225[degrees]C. The remaining two heating
zones were in the nozzle.
(b) The temperature was measured on the mould wall thickness by a type-K
thermocouple.

TABLE 3. The surface descriptors ([S.sub.q] and [[xi].sub.[parallel]])
of the smooth regions of three PP-2 moldings and those of the mold.*

Injection speed--mold
temperature--hydraulic holding
pressure                          [S.sub.q]  [[xi].sub.[parallel]]

Moldings (a)
  5 mm/s--30[degrees]C--30 bar    0.677      24.8
  120 mm/s--30[degrees]C--30 bar  0.076       8.1
  120 mm/s--60[degrees]C--60 bar  0.067      13
Mold (b)                          0.011      --

* Units in ([micro]m).
(a) Measurements using the Talystep Contact Stylus Profilometer.
(b) Measurement using the Wyko RST Plus Interferometer.

TABLE 4. The surface descriptors ([S.sub.q] and [[xi].sub.[parallel]])
of the fine regions of two PP-2 moldings and those of the mold.*

Injection speed--mold
temperature--hydraulic holding
pressure                          [S.sub.q]  [[xi].sub.[parallel]]

Moldings (a)
  5 mm/s--30[degrees]C--30 bar    4.3        95
  120 mm/s--30[degrees]C--30 bar  5.1        90
Mold (a)                          6.1        78

* Units in ([micro]m).
(a) Measurements using the UBM Optac 2000 Optical Profilometer.

TABLE 5. The surface descriptors ([S.sub.q] and [[xi].sub.[parallel]])
of the coarse regions of two PP-2 moldings and those of the mold.*

Injection speed--mold
temperature--hydraulic holding
pressure                          [S.sub.q]  [[xi].sub.[parallel]]

Moldings (a)
  5 mm/s--30[degrees]C--30 bar    31.1       330
  120 mm/s--30[degrees]C--30 bar  30.3       315
Mold (a)                          33.4       400

* Units in ([micro]m).
(a) Measurements using the UBM Optac 2000 Optical Profilometer.