Investigation on Properties Deterioration of GF (PA3200) Powder in Laser Sintering Process

This study is designed to investigate the Glass-filed GF polyamide (PA3200) GF, nylon 12 behaviour due to deterioration of the powder. For describing and understanding these causes to find systematic approach to improve the quality of parts and to reduce the orange peel texture. The thermal behaviour and melt flow rate (MFR) of the virgin and recycled were studied by differential scanning calorimetry (DSC), and melt flow rate testing (MFR). DSC results indicated that the three transitions temperatures were affected by reprocessing (number of cycles). MFR testing showed that the melt flow rate of glass filled (PA3200) powder was decreased by the temperature, time and reprocessing. The main problem related to this practice is how to choose the right amount of recycled powder to be blended with virgin powder, in order to obtain materials that have good properties, do not show significant variation from the virgin powder, without greatly deteriorate the properties of the final material with respect to the virgin one. However its quality was degraded due long hours expose at elevated temperature caused molecular chain structure and thermal properties have changed. For this reason and according to the EOS and 3D systems companies this investigation used 50% old blended with 50% virgin powder.


INTRODUCTION
"Laser sintering LS is rapid manufacturing technology in which a three dimensional object is created layer by layer from heat fusible powder materials with generated from CO2 laser" [1][2][3][4]. various polymer material, such as polystyrene (PS), polycarbonate (PC), and acrylonitrile butadiene styrene (ABS), ceramics, metal and polyamide (PA) using in SL process [5], where polyamide Duraform PA and Duraform GF polyamide of 3D System Company and PA 2200 and PA 3200 GF of EOS Company are suitable for functional parts where GF composite material consist of polymer and glass filled. Composite material can be defined as macroscopic combination of two or more distinct materials, having a recognisable interface between them [6].Most of the composite used are mainly characterised by very high specific stiffness and strengths, combined with an excellent resistance to the environment. Polymer composite can be either thermoplastic or thermoset the most commonly used reinforcements are glass and carbon fibbers, mineral and organic filler [7]. The applications of polymer composites have grown continuously and can be uses in transportation, building, electric, marine and aerospace industrial. "Using PA3200 (GF) material's, was used to improve dimensional stability, control shrinkage, reduce formation of micro cracks, improve surface finish and appearance and that the Castform material of 3D System Company and PrimeCast of EOS Company are suitable for investment casting" [8]. "Classification of LS technologies of composite material according to the binder mechanism most of these can be seen as liquid phase sintering of composite powder grains regarding the polymer as a binder and the glass particles as a structure material and basically it is a structure material that is fully molten, be it that its glass particles remain solid" [9]. Therefore, it is very important to obtain accurate prototype as quickly as soon possible, some complications that the LS process has commonly placed with the reuse of powder. Each time a part is built, there is loose material, which usually is sifted and reused to build other parts [10]. As the powder is reused, the parts which produced by SLS could be becomes poorer and the quality of the parts becomes imperfect such as rough surface, orange peel, poor properties and, LS can be also produced by recycled powder [11]. This paper examines glass filled GF composite material (PA3200) as a virgin powder (VGF) 50%, recycled powder 50% (RGF), 100% old powder and remixed 50%:50% powder to find the effect of the variation of temperature and time close to the process conditions of the LS build for making parts and an economic point of view.

Characterization of Powder Particle
Main characteristics of powders are the particle size (granulometry) and particle shape (morphology). Technological properties of powders (bulk density, flowability, surface area) as well as the potential areas of their application depend on those characteristics. The granulomerty of powders can be determined by different methods (sieve analysis, image analysis,) "size parameters (for example: average diameter, area or peimeter)describe a geometrical object independently of its shape. The morphology of a powder particle is characterized by description (spherical, angular, dendrite, dish-shaped, circular) the shape parameter characterizes mainly the shape, without considering the size" [12].

Effect of Molecular Weight on Viscosity
The most important molecular structure feature which influences the melt viscosity of a polymer is its molecular weight; melt flow rate (MFR) characterizes the melting viscosity of the polymer material. Thermoplastic polymer consists of many individual long chain molecules. The molecular weight of a polymer chain may be thought of as a measure of its size, but for essentially linear chains molecular weight is a measure of length. When these polymers are synthesised, the chain grow to different lengths. Hence, there will be an average chain length present. The average molecular weight Mw ( g mol -1 ) for these chains affects the melting viscosity 0 [7,8]. The high viscosity of polymer melts is chain entanglements which hinder the flow of individual chain over one another. Long chains are much more likely to be in a serious state of entanglement than short ones. The viscosity increase with increase molecular weight Mc [8]. Mathematically, the relationship between viscosity and molecular weight is

LS Process and Powder Properties
In the LS process there are processes parameters can be named into three categories: preheating, fusion and cooling rate control. The first two sets of parameters control part density, growth, and in-build curl; the last set controls growth and post-build curl. The feed and part heaters control preheating before delivering powder to the part bed with the roller, the feed powder must heat to 10C -15C below the melting point of the material at which the powder still flows freely and the part heater heats the powder to a temperature just a few degrees its melting point. Fusion occurs when just enough laser energy putting into the part to melt the powder. In general the LS process involves three stages of the polyamide: warm-up stage. during this stage, the system slowly increases the part bed powder temperature to 10C to 15C below its meting point; the build stage maintains the part bed temperature and uses relatively low laser power to melt the powder in the part boundary; the cooling stage allow the powder and the parts to cool slowly, process and recycling stages powder. During this stage, the system maintains the inert atmosphere to prevent oxidation (yellowing) of the powder and parts [13]. The material used in this work a glass filled (GF) nylon12 semi-crystalline powder, known as polyamide (PA3200) was supplied by EOS Company. The commercially supplied specification of the powder is particle size range of (25-92) m and an average particle size of 58 m, particles shape irregular, powder density 590 kg/m 3 and Melting temperature 185 C [14].

Blend Powders
After the parts manufactured by this type of EOS machine, loose glass filled GF powder can be sifted and reused in another build. Consistent recycling procedures are important in order to maintain consistent material properties. Mixing of powder is an important phase in this process, as uniformity of the mixture influences the flow behaviour of the sintering properties [15].The blends were prepared by mixing a virgin glass filled GF powder with the same glass filled GF after a processing cycle. For GF powder, combine the three types of powder: virgin, over flow and reused in a 1:1:1 ratio by weight [13]. Also the recommendation of EOS GmbH for GF (PA3200) to used refresh rate of new powder from 50% to 100% and 3D Systems recommended using 50% of refresh powder [17]. According to these the present work started by using 50% virgin powder and 50% old powder.

Melt Flow Rate Indexer
In order to understand the process conditions in various cases such as the effects of the temperature, the time and reprocessing during the LS process on the MFR and behaviour of powder after mixing before LS process and after the process on MFR. The trails were performed on a virgin and a loose powder to simulated the powders, since simulations were carried for the same conditions as the experiments, the samples were set in the Oven Heraeus Instruments Vacutherm, type 6060M in a maximum amount of Nitrogen atmosphere to prevent the powders from oxidation (yellowing) and heated at several different temperature beginning by 100 C, 120 C, 140 C, 160 C and 180 C for period time 15,25,50,75,100 and 150 hours at each temperature.
After these samples heated according to the temperature and time mentioned above, the (MFR) test indexer type Meltflier MT was used to determine the flow behaviour of the glass filed GF powder. The originally defined MFR method determines the mass flowing out in a defined time [16]. The MFR was selected as a criterion because the flow characteristics of a molten polymer are very sensitive to the changes in the basic polymer structure and its molecular weight [17]. The results are indicated in g/10min by extruding molten material from the die under preset conditions of temperature and load. According to ISO1133 standard a small of amount taken for testing (8-10 grams) under testing conditions 5.0kg and at temperature 275 C [18]. For melt mass flow rate, timed segments of extrude are weighed and extrude rate in g/10min and recorded [19] .In this kind of experiments, each samples were tested 6 times and MFR were measured then the average of them calculated.

3 Gel Permeation Chromatography
The GPC system used in these experiments to notice the influence of the time and temperature on the molecular weight and to determine the loose powder quality of different samples of glass filled of exposure time; virgin, 15, 54, 144 hours. A single solution sample was prepared by adding 5 ml of solvent to 15 mg of sample and warming at 40 C for two hours to dissolve. The aim of these experiments is to determine the molecular weight Mw and number of molecules Mn, [20,21].

The effect of the LS Temperature and Time on MFR
The experiments and simulations have been conducted in this work on the commercially supplied powders, known as polyamide (PA3200) GF. The specimens of the experiments were taken from the virgin (VGF), recycled 50% (RGF) and 100% old once used powders. After the preparation of specimens to be simulated, the samples of the powders were heated continuously. In general from the experimental measurements, the results after each build showed that the amount of material in grams that extruded in 10 minutes of this kind of powder decreased. And MFR was observed decrease after reprocessing and with each heat exposure, the curves in the figures below of samples showed the melt flow rate dramatically decreased after each exposure to heating at high temperature 160C and 180C rather than temperatures 100C,120C and140C and the values of these two high temperatures closely similar. For example, from the curves in the figure 2 it can be also noticeable the values of MFR greatly decreased at the 15 hours heating then the behaviour of the curves in this case decreased slightly until 100 hours heating, after that, the values of MFR starting to be constant as same as the values of 150 hours or remained limited change above 100 hours. When this observation was compared with the values of MFR for all the other samples in the figures 3 and 4 it was concluded that (PA3200) GF was very great affected by temperature, time and reprocessing, also indicating that the molecular weight increased at different temperatures and time induces reduction in the MFR. Indeed, due to powder exposure long time and RGF undergoes a sharp decrease in MFR also it related to increased molecular weight ( i.e. the length of chain effected on the MFR).

The Effect of Temperature and Time on the Molecular Weight and Viscosity
The GPC system used in these experiments to notice the influence of the time and temperature on the molecular weight and to determine the loose powder quality of different samples of glass filled of exposure time; virgin, 15, 54, 144 hours. The experiments showed that the Mw, Mn, MFR and viscosity for the sample mentioned above. MFR test used to evaluate the quality of these samples. It can be noticed from the figures 5, 6 and 7 there is a clear and large increase in molecular weight with time but with the biggest difference appearing to be between virgin powder and 15 hours exposure time. This behaviour was similar to that in the effect of time and temperature on the MFR it's decreased with the exposure time increase and MFR decrease was progressive at the beginning of the exposure time but more pronounced at the 15 hours. It was attributed to the molecular weight. Virgin powder with the highest MFR is the lowest molecular weight value and viscosity and the lowest sample was 150 Hours long exposure time. However the viscosity determined from the MFR measurements increased significantly as showed in figure 6. This indicates that the degradation affected mainly by exposure time required running the process.

The Effect of The Continuous and Cycling Heating on The MFR
In the following experiments, the samples were heated at two different temperatures 160 C and 180 C, two methods, continuous heating and cycling heating, the former were heated the samples continuously at several time then taken out and measuring melt flow rate, the latter the samples were heated at the same period time and cooled to the room temperature then they were returned to the oven after measuring MFR, the time for both method was 15,25,50,75,100 and 150 hours., as seen directly by these examples in figures 8, 9 the melt flow rate after processing steps for virgin (VGF) and recycled (RGF) blends of glass filled GF (PA3200), the points of the curves are related to the experimental values. It can be observed that the MFR of virgin and of the recycled 50%50% powder slightly different or fairly similar for both methods continuous as a reference and cooling to the room temperature then return to the oven as cycling. Although the data and experimental conditions were used the same, even the temperature was different, the decrease in MFR was remarkable of the reprocessing at 160 C and 180 C. MFR at 160 C curves in the figures 8, 9 showed behaviour of reference samples (continuous heating) slightly higher by one at the virgin powder VGF, recycled 50%50% was very similar and 100% old once used the highest value 2.9 g/10min compare to that samples of cycling heating.
Moreover MFR increased slightly between both of the methods, from a value of 0.2 g/10min for the 1X recycled material to 0.5 g/10min. Also noticed that the temperature 180C the same behaviour was noticeable even the MFR values were lower. Thus, the appears continuous heating or cycling heating to be less effecting and not the main parameter controlling reprocessing and deterioration of glass filled GF powders on the MFR of glass filled powder. In addition to the MFR the colour GF was effect in both methods of simulated at temperature 180 C the originally white powder showed change to dark.

0 CONCLUSIONS
The quality of polyamide (PA3200) GF powder before LS process and sintering hours influences the deterioration on the unsintered powder and surface finish of the parts produced. The lower melt flow rate value means the higher deterioration. The MFR of PA3200 (GF) showed a more much rapid decrease with the number of cycles than that of PA., and was more pronounced in recycled 50% and old powders.