Influence of Vacuum on the Porosity and Mechanical Properties in Rotational Molding.
Rotational molding is a polymer processing technology for the creation of seamless, hollow parts, which excels by a high degree of design flexibility and the ability to create parts with large dimensions like storage tanks, casings for engines or kayaks . Relatively long cycle times and porosity are two limitations of the process, which restrict each other contradictorily . On the one hand, the creation of pore-free products requires long sintering times and is thus only economical in small- and medium-scale production . On the other hand, the reduction of cycle time results in an increase in porosity , which can affect mechanical properties  and surface qualities . The feasibility of reducing the porosity by applying positive pressure was already shown for rotational molding . The use of vacuum pressure gives the opportunity to apply a pressure difference and overcome the danger of burst and oxidation effects of the material. This could open up new mold manufacturing considerations as well as possibilities of material selection. Vacuum integration shows high potential to reduce pores as shown in oven trials by Lohner and Drummer  but was not yet evaluated in rotational molding to its full extent. In this article, an experimental setup is developed to investigate the influence of vacuum pressure and holding time on the porosity, surface appearance, and mechanical properties.
The rotational molding process can be divided into four elementary steps . In the first step of the process, polymer powder or liquid is introduced into a hollow mold. After closing, the mold is rotated along two axes and heated up in the second step. The polymer powder starts to melt in the rotating tool and adheres to the mold's wall. Thus, a porous layer is created at the inner surface of the mold, which densities to a closed layer of molten polymer. In the third step, the rotation is maintained and the mold is cooled, which results in a solidification of the polymer. The mold is opened and the part can be demolded in the fourth step. The processing steps of rotational molding are illustrated in Fig. 1.
The number of materials that are feasible for processing in rotational molding is limited due to economic and technical aspects. In addition to an efficient production process, the material has to fulfill additional prerequisites like good powder flow, thermal stability, and moderate viscosity. Polyethylene is with about 90% by far the most utilized polymer in rotational molding. Several other materials like Polyvinylchloride, polycarbonate, and polypropylene can also be used in rotational molding, but have a lesser economic impact [2, 7].
The sintering behavior of the powder in the heating phase is a crucial aspect of the rotational molding behavior and is responsible for the creation of air inclusions, as investigated by Kontopoulou and Vlachopoulos . Figure 2 shows the melting and densification schematically. When heating the mold, the powder inside the mold also heats up and adheres to the wall of the mold, when a critical temperature is reached (a). With further heating time, the particles plasticize due to the heat conduction of the mold, which enables further particles in the powder bed to stick to the plasticized surface (b). The lowest layer melts completely (c). Air inside this layer is trapped and forms air inclusions. By this procedure, the part is built up successively. When the whole powder adheres to the wall of the mold and is completely molten (c-e), additional holding time is needed for the air inclusions to dissolve and to create a nonporous part (f). 
Due to this sintering process, bubbles in the melt cannot be avoided. However, powder shape and size influences the initial bubble size and thus the density of the part  More regular shaped particles result in a higher initial bulk density . Larger particle dimensions lead to a less efficient sintering process . An increased proportion of fine particles in the powder lead to an increased number of bubbles of smaller dimensions . Besides the initial size of the bubbles, different influences on the bubble dissolution rate were investigated, including the temperature of the melt  (also overheating ) and other material related properties like the surface tension and the air concentration . The main phenomenon for the shrinking and removal of bubbles are pointed out by Xu and Crawford :
* gas diffusion at the surface of the bubble,
* diffusion of the dissolved gas from the bubble into the polymer melt and
* buoyancy forces of the bubble.
It was shown for PE that the buoyancy forces are insignificant because of the high viscosity of the rotational molding material . Two main effects are triggered by the driving force of the pressure difference between the air in the bubble and the surrounding melt, resulting both in a concentration gradient, according to Gogos . On the one hand, the surface tension, which causes a solution of the gas molecules in the surrounding melt, is responsible for the bubble shrinkage. On the other hand, the dissolution of trapped air is caused by the under-saturated surrounding melt. Due to the long heating times in standard rotational molding, the polymer is probably almost saturated. At saturated melts, the surface tension effect, which has a minor influence on the bubble dissolution time compared to the driving force of the unsaturated melt, prevails. Long bubble lifetimes are the consequence. 
The time dependency of the bubble dissolution is one reason for the relatively long cycle times in rotational molding. The bubble removal takes approximately 40% of the cycle time in conventional rotational molding processes . This is a major disadvantage for the process , as it reduces the number of pieces that can potentially be produced. If the cycle time is too short, air inclusions remain in the part which can lead to impaired mechanical properties . Moreover, too short cycle times result in a higher surface roughness due to time and temperature depending smoothening effects [5, 6].
Due to these restrictions, various researchers have investigated the potential of pressure in rotational molding. In order to reduce porosity, the positive pressure is applied after the polymer is completely molten . Bubble shrinkage due to compressed air in the inclusions is caused according to Boyle's law . Besides this, the solubility of the air increases. An increase in pressure enhances the air solubility in the melt due to the concentration gradient driving force of the under-saturated melt . Applying positive pressures, this driving force has the largest effect on the bubble dissolution time compared with negligible surface tensions effects (presuming typical bubble sizes for rotational molding) .
The influence of positive pressure in rotational molding polyethylene was also examined by Spence and Crawford . The porosity of the specimen decreases, when the pressure is introduced after the polymer is molten. Moreover, tensile strength (+5%) and impact strength (+25%) increases, surface quality is improved and shrinkage reduced. The time interval of pressure application influences these correlations. The mentioned trends were substantiated by further investigations [11, 13-15].
The positive effect of vacuum in rotational molding was mentioned in previous investigations: Single trials were conducted in rotational molding, which show the potential of vacuum [4, 16]. In Ref.  it was shown on a heating plate, that the application of vacuum can be beneficial for the dissolution of air inclusions, if the vacuum is released while the material is in viscous state. Increasing the temperature results in an increase of bubble diameter at first. Due to the release of the vacuum and the consecutive maintenance of ambient pressure, the bubbles shrink and collapse instantly. Crawford and Throne  note that no hard vacuum is necessary to obtain beneficial effects in rotational molding. This was also examined in oven trials by Lohner and Drummer : a distinct reduction of porosity of more than 90% was observed, when applying a pressure difference of 100 mbar (from ambient pressure to 900 mbar) during melting. At a pressure difference of 300 mbar, a quasi-pore-free specimen was created. With the setup, a reduction of the melting and densification process time of 70% could be achieved. However, systematic investigations to understand the interaction of vacuum level, viscosity, and time on consolidation behavior have not been carried out yet.
MATERIALS AND METHODS
In this study, the linear high-density polyethylene powder UP203 is utilized for the rotational molding trials, provided by Ultra Polymers (Lommel, Belgium). It was developed for the rotational molding process with heat and UV stabilization. The density of the material is 0.9395 g/[cm.sup.3] according to the material manufacturer.
Rotational Molding Experiments
Uniaxial rotational molding experiments were performed with a sealed cylindrical mold with a diameter of 100 mm and a length of 200 mm and a rotating speed of 45 [min.sup.-1]. Fig. 3a. The outer mold surface was heated by two infrared heaters (Optron GmbH, 2 kW maximum power each, Garbsen, Germany) and cooled by compressed air. The temperature of the outer mold's surface was captured by a pyrometer, which was used for process control. The temperature difference between inner and outer mold surface, captured in preliminary static tests, was negligible. The process started at a temperature of 55[degrees]C with the heating phase, as shown schematically in Fig. 3b. When the temperature of 180[degrees]C was reached, the temperature was held for a defined time [t.sub,h]. After the holding time, the mold was cooled down to a demolding temperature of 85[degrees]C. Vacuum was applied in this process, utilizing a rotary joint. At the beginning of the heating phase, the mold was evacuated, Fig. 3b. The vacuum level [DELTA][p.sub.v] is displayed as the magnitude of the pressure difference relating to ambient pressure (1,000 mbar). Before the cooling phase started, the vacuum was released and ambient pressure ensured. For every trial, a repetition rate of three was utilized.
A variation of the vacuum level was conducted with this setup, Table 1. Trials with vacuum pressure differences [DELTA][p.sub.v] of 0 mbar (ambient pressure), 250, 500, and 750 mbar were performed. The other processing parameters were held constant to isolate the effect of the vacuum level. Furthermore, the influence of the holding time was investigated at absolute vacuum pressure deltas of 0 and 750 mbar, Table 1. Trials with holding times of 0, 150, 300, and 450 s were conducted.
The melting behavior of the material was characterized by differential scanning calorimetry (DSC) employing a Q2000 calorimeter (TA Instruments, Delaware). The experiments were conducted according to DIN EN ISO 11357 with a defined cooling and heating rate of 10 K/min. The rheological behavior was measured by a plate rheometer (AR 2000). The measurement was conducted with an oscillating frequency of 0.1 Hz and a strain of 0.5%. The particle size distribution of the powder was derived by a Morphologi G3 particle characterization system (Malvern Instruments, Malvern, Great Britain) comprising the diameter of 36,470 particles. The particle shapes were illustrated exemplarily using a scanning electron microscope (SEM) Gemini Ultra-Plus (Carl Zeiss AG, Oberkochen, Germany).
In order to evaluate the porosity, density measurements were conducted applying the buoyancy method (DIN EN ISO 1183). The following equation was used for the calculation of the porosity P:
P = ([[rho].sub.s]/ [[rho].sub.p] - 1) / ([[rho].sub.a]/ [[rho].sub.p]-1) (1)
with the measured density of the specimen [[rho].sub.s] the density of the polymer [[rho].sub.p]. and the density of the air [[rho].sub.a]. An air density of 0.1293 g/[cm.sup.3] is assumed for the ambient testing temperature of 23 [degrees]C. The density of the polymer pp is 0.9395 g/[cm.sup.3] according to the material manufacturer. With three rotational molding trials for each parameter combination and three density measurements, the total number of samples for each parameter combination was nine. For visualizing the gas inclusions, transmitting light microscopy images of the cross section of rotational molded samples (orthogonal to rotation direction) were prepared by thin-cutting technology (10 [micro]m) and recorded applying a polarization filter. The surface structure was illustrated and quantified by laser scanning microscopy using an OLS4000 (Olympus, Tokyo, Japan). An image of 7,150 mm X 9.507 mm was computed by capturing a grid of 12 single images. To calculate the waviness, a cutoff wavelength of [[lambda].sub.c] = 0.2 mm (low-pass) and [[lambda].sub.f] = 8.0 mm (high-pass) was applied. The waviness Wa was determined by calculating the arithmetical mean deviation of the filtered profile (3D measurement). The influences of the vacuum level and the heating time on the mechanical properties of the part were investigated by tensile testing according to DIN EN ISO 527 using an universal testing machine type Zwick 1,465 (Zwick Gmbh &Co. KG, Ulm, Germany) and a test speed of 30 mm/min. Therefore, tensile bars with a width of 10 mm and a thickness of 2 mm were milled out of the cylindrical part in axial direction. The test length was 70 mm. With three rotational molding trials for each parameter combination and five tensile tests, the total number of samples for each parameter combination was 15.
RESULTS AND DISCUSSION
The melting behavior is characterized by DSC. The melting peak of the second heating occurs at a temperature of 127[degrees]C; the melting enthalpy is 165 J/g. During cooling, the crystallization starts at 116[degrees]C and has its peak temperature at 114[degrees]C. Assuming that the mold temperature is equal to the material temperature in the mold, the processing temperature of 180[degrees]C is sufficient to melt the powder. The viscosity of the material, captured by rotational viscosimetry, decreases with increasing temperature, Fig. 4. At the processing temperature of 180[degrees]C, a viscosity of 1947 Pa s occurs. The used strain of 0.5% and frequency of 0.1 Hz results in a maximum shear rate of 0.14 [s.sup.-1] and can thus be compared to the quasi zero-shear rate, which is present in rotational molding. Therefore, the results gathered by rotational viscosimetry can be transferred to the rotational molding process.
Referring to the literature , the upper limit for diffusion of trapped air in rotational molding appears to be approximately 3,000 Pa s to 4,000 Pa s. According to this estimation, diffusion is a relevant factor for the bubble dissolution of the material and the process parameters used.
Different particle sizes can be observed by SEM and particle analysis. Figure 5a shows the SEM image of the powder. The particles have an inhomogeneous shape and occur in different particle sizes, which are quantified by the particle characterization system, Fig. 5b. The smallest particle detected has a diameter of 3.7 [micro]m, the biggest a diameter of 766.6 um. Of the powder volume, 10% indicates particle diameters smaller than 70.78 [micro]m, of 50% the particle diameter is smaller than 238.9 [micro]m, and of 90% smaller than 627.2 [micro]m. A broad distribution of particle sizes compensates the low packing density of big particles and leads to smaller bubble sizes in the melt . However, the irregular shape of the particles decreases the sintering efficiency .
Rotational Molding Experiments
The influence of vacuum pressure and holding time on the porosity of the samples is examined qualitatively by transmitting light microscopy, Fig. 6. At ambient conditions ([DELTA][p.sub.v] = 0 mbar), air inclusions occur at a holding time of 0 s. At a holding time of 450 s, bubbles can still be detected in the cross section. Applying a pressure difference of 750 mbar, no visually recognizable inclusions remain in the samples, independent of the holding time. Regarding the inner surface of the part, which is displayed in the upper section of the images, a wavy structure occurs for both pressure levels at a holding time of 0 s. A smooth surface is achieved at a holding time of 150 s.
The influence of vacuum pressure on the bubble dissolution, as shown qualitatively in Fig. 6, is characterized quantitatively by the porosity (according to Eq. 1), Fig. 7a. At [DELTA][p.sub.v] = 0 mbar a porosity of 1.68% occurs. Increasing the pressure difference results in a severe drop in porosity to 0.22% at [DELTA][p.sub.v] = 250 mbar. There is a further minor decrease in porosity of 0.06% with an increasing pressure difference at [DELTA][p.sub.v] = 750 mbar, indicating a quasi-pore-free part.
Figure 7b compares the porosity (according to Eq. 1) at [DELTA][p.sub.v] = 0 mbar and [DELTA][p.sub.v] = 750 mbar for the variation of the holding time. At ambient pressure ([DELTA][p.sub.v] = 0) a decent decrease of porosity from 1.99% at [t.sub.h] = 0 s to 1.61% at [t.sub.h] = 450 s can be observed. Utilizing a pressure difference of 750 mbar results in quasi-zero porosity for holding times between 150 s and 450 s. Minor average porosity of < 0.3% occurs at this vacuum pressure and a holding time of 0 s.
It is supposed that due to the absence of vacuum pressure in the process at [DELTA][p.sub.v] = 0 mbar, the melt is saturated. Thus, the predominant effect for bubble dissolution is the surface tension . The bubble dissolution based on surface tension is more time intensive than the mechanism based on the concentration gradient , which is substantiated by the existence of bubbles at a holding time of 450 s, see Figs. 6 and 7b. The dissolution of air inclusions by releasing the vacuum (after a melt layer is formed) is probably caused by two main effects. Increasing the pressure to ambient pressure results in a shrinkage of the bubbles due to Boyle's law. Furthermore, due to the pressure difference, the melt probably becomes unsaturated resulting in a concentration gradient, which accelerates diffusion. Figures 6 and 7a indicate that a pressure difference of 250 mbar is sufficient to create a quasi-pore-free part. This statement is in accordance with Lohner and Drummer, who created a quasi-pore-free part in oven trials at a pressure difference of 300 mbar . At [t.sub.h] = 0 s and [DELTA][p.sub.v] = 750 mbar, a porosity of 0.3% is measured, whereas no air inclusions are visible in the corresponding microscopic images, Fig. 6. One possible reason is the impact of the surface topology on the density measurement. It can be concluded for the conducted parameter variations at the applied working temperature that the pressure difference has a bigger influence on the porosity than viscosity and surface tension effects at ambient pressure. A variation of processing temperature is planned for future investigations to compare the effects of viscosity changes and pressure differences quantitatively.
As recognizable in the light microscopy images, a waviness occurs at holding times of 0 s. The structure of the inner surface of the specimen is visualized and quantified by laser scanning microscopy, Fig. 8. At ambient pressure ([DELTA][p.sub.v] = 0) and a holding time of 0 s, the highest waviness of 71 [micro]m appears. The waviness decreases to 22 [micro]m at 150 s holding time and remains on a low level with a further increase in the holding time. Referring to a pressure difference of 750 mbar, the same trend emerges: a high waviness of 62 [micro]m occurs at [t.sub.h] = 0 s, which is reduced with longer holding times to 27 [micro]m at [t.sub.h] = 150 s resp. 24 [micro]m at [t.sub.h] = 450 s.
Comparing the waviness of [DELTA][p.sub.v] = 0 mbar and [DELTA][p.sub.v] = 750 mbar, no correlation between pressure difference and surface quality can be derived. It can be stated, although, that the reduction of cycle time by means of vacuum pressure is restricted due to the time depending smoothening effect of the inner surface. A smooth surface is desirable to avoid groove effects and to create homogeneous part thicknesses. In terms of application, a smooth surface is a basic requirement, for example, for the flowability of gases and liquids in a storage tank.
Tensile tests were carried out in order to correlate the influences of vacuum with varying pressure levels and different holding times with resulting mechanical properties. Figure 9a shows the tensile strength to yielding of the specimen in dependency of the vacuum pressure difference at a holding time of 300 s. At ambient pressure ([DELTA][p.sub.v] = 0 mbar), a tensile strength of 16.6 MPa occurs. Increasing the pressure difference to 250 mbar, the tensile strength rises to 17.4 MPa. A further increase in [DELTA][p.sub.v] results in a minor increase of tensile strength to 17.6 MPa at [DELTA][p.sub.v] = 500 mbar, resp. 17.8 MPa at [DELTA][p.sub.v] = 750 mbar, which is, however, in the range of the standard deviation.
The tensile strength in dependency of the holding time at [DELTA][p.sub.v] = 0 mbar and [DELTA][p.sub.v] = 750 mbar is illustrated in Fig. 9b. Referring to ambient pressure ([DELTA][p.sub.v] = 0 mbar), the lowest tensile strength of 15.8 MPa appears at 0 s holding time. The tensile strength rises to 16.9 MPa, applying a holding time of 150 s. A further increase in [t.sub.h] shows no clear trend for the tensile strength. A similar behavior is detected for a pressure difference of 750 mbar at slightly higher tensile strengths. The tensile strength rises from 16.5 MPa at [t.sub.h] = 0 s to 17.4 MPa at [t.sub.h] = 150 s and remains on the same level, with increasing holding time. However, due to the high standard deviation at 0 s holding time, no strict relation can be derived here. The lower level of tensile strength for the specimens which were manufactured without holding time (compared to longer holding times) at pressure differences [DELTA][p.sub.v] = 0 mbar and [DELTA][p.sub.v] = 750 mbar can be attributed to the surface structure. As shown in Fig. 8, a high surface waviness is present at [t.sub.h] = 0 s. When applying a tensile load the waviness acts like a surface error and can lead to notch effects. Regarding the variation of pressure at a holding time of 300 s, Fig. 9a, the beneficial effect of vacuum pressure on the tensile strength can be explained by the absence of pores at [DELTA][p.sub.v] [greater than or equal to] 250 mbar and thus a reduction of errors in the material.
Apart from the tensile strength, the Young's modulus, and the elongation-at-break show similar dependence on the parameter variations, Fig. 10. In dependence of the vacuum pressure difference, the poorest attributes can be observed for the specimens produced without vacuum application. Young's modulus and elongation-at-break increase slightly for [DELTA][p.sub.v] = 250 mbar and then remain on a comparable level for increased vacuum pressure differences. Since the porosity hardly changes in dependence of the vacuum pressure difference in the range from 250 mbar to 750 mbar, both the constant values of the modulus of elasticity and the resulting elongations at break, which also maintain a comparable level, represent a conclusive result. Figure 10b shows furthermore, that better mechanical properties are always achieved when applying vacuum, independent of the holding time. These changes can be attributed to the eliminated part defects in form of inclusions. Nevertheless, the surface waviness of the specimens seems to have a bigger effect on the elongation at break. This is shown by the fact, that independently of the vacuum pressure, a comparatively strong increase of the mechanical properties is achieved when comparing the results of components manufactured at 0 s holding time with components manufactured at holding times [greater than or equal to] 150 s, Fig. 10b. This effect is not apparent when considering the modulus of elasticity, because it is a material constant.
Comparison to Positive Pressure
In this work, similar correlations between the application of vacuum pressure and positive pressure on the morphology and mechanical properties of rotational molding parts could be achieved. In the first case, a vacuum is created before sintering and the pressure is increased to ambient pressure after melting. In the other case, the pressure is increased after melting from ambient pressure to elevated pressures. Both techniques include the application of a pressure difference after melting the polymer. It remains unclear whether it can therefore be assumed that the two process variants rely on the same mechanisms. More detailed investigations of the physical processes that take place in a vacuum application are intended to substantiate the findings. In Ref.  it was shown that the application of a positive absolute pressure of 1,500 mbar after the polymer has melted successfully removes air inclusion in the rotationally molded part. Similar results could be achieved in this work by applying vacuum. In both cases, the magnitude of the pressure difference is a main factor of bubble dissolution, affecting dissolution effects and shrinkage due to Boyle's law. The results of the porosity measurements permit the conclusion that a soft vacuum is sufficient to produce a quasi-pore-free part, Fig. 7.
In this work, a correlation between vacuum pressure difference and mechanical properties is shown. The beneficial influence of applying a positive pressure in rotational molding on the mechanical properties was also shown by Spence and Crawford. . By reducing component defects in the form of pores, an improvement in tensile strength of 5% was achieved. Similar results can be demonstrated in this work, with an increase in tensile strength of 4.8%, Fig. 9a ([DELTA][p.sub.v] = 0 mbar to [DELTA][p.sub.v] = 750 mbar).
Although positive pressure and vacuum application affect the rotational molding process in a similar way, the utilization of vacuum can be favorable. By applying vacuum, the danger of explosion can be bypassed, which allows advantages in mold design. Moreover, materials with vulnerability to oxidation can be processed due to the absence or reduction of oxygen.
In this article, an experimental setup is used to investigate the influence of vacuum and holding time on the morphology and mechanical properties in rotational molding. The mold is evacuated at the beginning of the process and vented to ambient pressure after the material has melted completely. For a holding time of 300 s, density measurements and light microscopy examinations have shown that applying a vacuum pressure creates quasipore-free samples with increased mechanical properties-compared to processing at ambient pressure. An influence of the vacuum level on the surface quality cannot be detected. A comparative analysis of the process with different holding times, with and without vacuum, shows a clear potential of the vacuum in regard to cycle time reduction. At ambient pressure, air inclusions are still present after a holding time of 450 s. Application of vacuum results in quasi-pore-free samples without the need of additional holding time. However, without the additional heating time, a wavy uneven surface occurs. Exemplarily, it could be shown that a smooth surface is achieved for an applied holding time of 150 s. This correlates with a cycle time reduction of 67%. Depending on the desired surface quality, the cycle time may be even further reduced.
The application of vacuum in rotational molding has similar effects on the morphology and mechanical properties as positive pressure. However, vacuum can be favorable as it reduces the danger of burst and oxidation effects. In future investigations, the variation of further process and material parameters like the holding temperature or the particle size is foreseen to increase the understanding of the process on the one hand and exploit the potential of vacuum on the other hand in order to minimize cycle time.
Lukas Vetter, Jannik Werner (iD), Michael Wolf (iD), Sebastian Hertie, Dietmar Drummer Friedrich-Alexander-University (FAU), Institute of Polymer Technology, Am Weichselgarten 9, 91058 Erlangen, Germany
Correspondence to: J. Werner; e-mail: email@example.com
Contract grant sponsor: Horizon 2020. contract grant sponsor: European Union.
Published online in Wiley Online Library (wileyonlinelibrary.com).
The authors want to thank the European Union's Horizon 2020 research and innovation program for the funding as well as Ultra Polymers (Lommel, Belgium) for the provision of the material used.
[1.] G. Beall, Rotational Molding: Design, Materials, Tooling, and Processing, Hanser, Munich (1998).
[2.] R.J. Crawford and J. Throne, In Rotational Molding Technology, Plastics Design Library/William Andrew Publ, Norwich, NY (2002).
[3.] M. Lohner and D. Drummer, J. Polym. Eng., 35(5), 481 (2015).
[4.] A. Spence and R.J. Crawford, Proc. Instrum. Mech. Eng., Part B., 210, 521 (2016).
[5.] M. Kontopoulou and J. Vlachopoulos, Polym. Eng. Sci., 39, 1189 (1999).
[6.] M. Lohner and D. Drummer, J. Polym. Eng., 37(4), 411 (2016).
[7.] R.J. Crawford and M.P. Kearns, Practical Guide to Rotational Moulding, Vol. 2, iSmithers Rapra Publishing, Shawbury, UK (2012).
[8.] M. Kontopoulou and J. Vlachopoulos, Polym. Eng. Sci., 41(2), 155 (2001).
[9.] L. Xu and R. Crawford, J. Mater. Sci., 28(8), 2067 (1993).
[10.] A. Greco and A. Maffezzoli, J. Appl. Polym. Sci., 92(1), 449 (2004).
[11.] A. Spence and R.J. Crawford, Polym. Eng. Sci., 36(7), 993 (1996).
[12.] C. Bellehumeur and J. Tiang, Polym. Eng. Sci., 42(1), 215 (2002).
[13.] G. Gogos, Polym. Eng. Sci., 44(2), 388 (2004).
[14.] M. Oliveira, M. Cramez, C. Garcia, M.P. Kearns, and E. Maziers, J. Appl. Polym. Sci., 108(2), 939 (2008).
[15.] R.J. Crawford, A. Spence, M. Cramez, and M. Oliveira, Proc. Instrum. Mech. Eng., Part B., 218(12), 1683 (2004).
[16.] L. Bergamo, P. Spa, and R. J. Crawford, Measurement and Control of Pressure Inside Rotational Moulds, https://rotoworldmag. com/measurement-and-control-of-pressure-inside-rotational-moulds/ (accessed March 7, 2018).
Caption: FIG. 1. Rotational molding process according to Ref. , (a) Filling; (b) heating; (c) cooling; (d) demolding.
Caption: FIG. 2. Sintering behavior in rotational molding according to Refs. [3, 8]. (a) Powder adheres to the mold wall; (b) further particles stick to the plasticized particles; (c) complete melting of the first layer; (d) whole powder is plasticized or molten; (e) whole powder is molten; (f) pore free part.
Caption: FIG. 3. (a) Sketch of experimental setup of the mold and (b) schematic pressure and temperature profile utilized at the rotational molding experiments.
Caption: FIG. 4. Viscosity of polyethylene UP 203 captured by rotational viscosimetry.
Caption: FIG. 5. (a) Powder particles of UP 203 captured by scanning electron microscopy and (b) particle-size distribution of UP 203.
Caption: FIG. 6. Transmitting light microscopy images of the cross section of rotational molding samples with holding times of 0, 150, and 450 s at 0 mbar and 750 mbar vacuum pressure difference.
Caption: FIG. 7. Porosity derived from density measurements; (a) Dependency of the vacuum pressure level at a holding time of [t.sub.h] = 300 s; (b) dependency of the holding time for vacuum pressure differences of 0 mbar and 750 mbar.
Caption: FIG. 8. Laser scanning microscopy of the inner surface of the specimen at holding times of 0, 150, and 450 s for pressure differences of 0 mbar and 750 mbar.
Caption: FIG. 9. Tensile strength (to yielding) in dependency of the vacuum pressure level at a holding time of 300 s (a) and in dependency of the holding time (b) according to the standard DIN EN ISO 527.
Caption: FIG. 10. Young's modulus and elongation in dependency of the vacuum pressure level at a holding time of 300 s (a) and in dependency of the holding time (b) according to the standard DIN EN ISO 527.
TABLE 1. Overview of the parameters varied in the test series: Variation of vacuum pressure difference [DELTA][p.sub.v] and holding time [t.sub.h]. Variation Pressure difference Holding time [DELTA][p.sub.v] (mbar) [t.sub.h] (s) Pressure 0, 250, 500, 750 300 Holding time 0 0, 150, 300, 450 Holding time 750 0, 150, 300, 450
|Printer friendly Cite/link Email Feedback|
|Author:||Vetter, Lukas; Werner, Jannik; Wolf, Michael; Hertie, Sebastian; Drummer, Dietmar|
|Publication:||Polymer Engineering and Science|
|Date:||Aug 1, 2019|
|Previous Article:||Anionic Copolymerization of Nylon 6/12: A Comprehensive Review.|
|Next Article:||Influence of Screw Configuration and Processing Temperature on the Properties of Short Glass Fiber Reinforced Polypropylene Composites.|