In recent years, fused deposition modeling (FDM) has become a widely used technology in rapid prototyping process. Typically polymer materials used are acrylonitrile butadiene styrene (ABS) and polycarbonate (PC). The polymer is fed into equipment as a 1.6mm diameter, which is heated and extruded onto a building platform. The ABS filament sold by the equipment suppliers is very expensive at aroundï¿¡270 per kg [1]. Also, during the FDM process, some of this material is used as build supports and which is simply thrown away after being removed from the parts. This project aims to use alternative materials for the FDM process.

1.2 Previous Work of Project

This project is followed by a summer 2009 project that Yifei Zhang did. In Zhang’s project, one grade of virgin ABS has similar melt and flow properties to the benchmark material, but due to the limitations of laboratory equipment and project funds, Zhang cannot get the result. Zhang also try to use WEEE ABS as a potential source, but the degree of purity (98%) could not be accepted in FDM operation. In addition, a nylon strimmer line was found which had the suitable filament diameter, but it also failed because of the viscosity of the materials.

1.3 Objectives of Project

After summarize the Zhang’s experience, the target of project is to identify the cheap alternative polymeric materials to ABS. Before the project, a further polymer understanding of the melt and flow properties should be reviewed. Some problems Zhang met need to be solved to successfully produce the product. Some experiments, which Zhang did, need to be repeated to have a better understanding of this project. Some trial materials are aimed to produce the products, and then the properties of product are evaluated.

2. Literature Survey

2.1 Why use rapid prototyping methods?

Rapid Prototyping (RP) is defined as ” the use of a computer aided design (CAD)-based automated additive manufacturing process to construct parts that are used directly as finished products or components”[2]. Generally, it is a quick way to turn a virtual design to a real model. There are many purposes for a prototyping, for example, validate system specification or deliver early proof of concept or create a market demo [3]. The best reasons to rapid prototyping is to save time and save money. A complete product design cycle is very long, and there are a lot of trials and errors in the processing. The rapid prototyping processing can be used to modify the trial and debug the prototype in a short time. In addition, the resources can be saved by producing reduce scale model.

2.2 Rapid prototyping equipment

With the development of modern science and technology, techniques of rapid prototyping technology increasingly diversified, such as stereo lithography (SL), laminated object manufacturing (LOM), selective laser sintering (SLS), selective laser melting (SLM), fused deposition modeling (FDM), ink jet printing (IJP), 3-D printing (3DP) [4]. There are three categories according to the raw material used in these processing. They are liquid-based system, powder-based system and solid-based system [2] [3]. FDM belongs to solid-based RP processes. The product made by FDM operation can be used as trial tests or for further studies, which reduces the risk of designing mistakes.

2.2.1 Stratasys FDM System

Fused Deposition Modeling (FDM), which invented by Stratasys Inc, is used in the project. Generally, FDM is a typical PR process, which contains a computer controlled, and micro-extrusion system. In the process, the polymer is fed into the equipment as a 1.6 diameter filament, which is heated and extruded through a die as a 0.5mm fluid filament. By lay down of the melt filament from nozzle, certain thickness of 2-D layers structure is formed, then by repeating the process for the second layer, repeat this process layer by layer, then a complex 3-D structure is formed. As can be seen in Fig 1, the processing route follows the design, forming and finishes these three steps.

In the route of this manufacture, pre-process CAD file is designed by FDM Quickslice software and saved as STL format [6], and then during the manufacture part, the thermoplastic material is extruded and formed layer by layer on precise paths. Remove the support structures then the final product is completed.

Extrusion process is a key step in the FDM manufacturing [Fig 2]. Generally speaking, the polymer filament is in solid state when feeding into the FDM machine. The filaments of molten thermoplastic are extruded from a heated nozzle. The process lays down both support material filament and build material filament for one layer at a time. The diameter of the filament is about 1.67mm, and the range of diameter of different tips is from 0.025mm to 0.4mm.

Functional prototyping is used via an additive fabrication system. Manufacturing tools and End-Use parts also can be produced with FDM system. There are many advantages including streamlining the development process, reducing the cost and getting to the market faster [7].

The disadvantage is obvious, the price of cartridge is very expensive, and ABS build (including the breakaway materials) is £275.00 per kg. So find the cheap alternative material is the target of the project.

2.2.2 Mesostructure of FDM ABS [8]

There are two kinds of mesostructure during the FDM processing [Fig 3]. The aligned mesostructure is the standard model of the machine. Another skew mesotructure uses the alternating “jogs” of length 0.5(W+g) in the ProEngineer drawing [Fig 4]. A correction in the nozzle height setting, δ z (see equation 1), using the z-direction shrinkage factor, is set in the QuickSlice program.

The void density on the material plane i is defined as see equation 2). The extent of circumferential fiber-to-fiber bonding is quantified using a cross-section average of the fiber-to-fiber interface bond length densities. For a single fiber, it is defined as: (see equation 3)

According to Jose, James and John, the same filament materials (ABS P400) and the same serial FDM machine were used to test. So the data have the reference value.

As can be seen in Fig 5, the lowest void density value is the skewed configuration structure. From the chart above (Fig 5 Fig 6 and Fig 7), Fiber gap (g) and flow rate are two key factors on mesostructure while temperatures (Te and TL) had very little influence. So for the best properties, minimum void and maximum bond length are needed.

2.3 Alternative material

2.3.1 Standard Material

The current materials using for the FDM manufacturing process are supplied by Stratasys Inc, which could be ABS plastic, PC and wax [9]. The P400 ABS, whose main mechanical properties can be clearly know in Fig 8, bought from the machine vendor was used as benchmark this project, provided as white filament with a diameter of 1.67mm and a ± 0.02 mm tolerance, measured by laboratory’s calipers.

2.3.2 Recycled ABS

Recycled ABS, I think, still is the most ideal choose in this project. The recycled WEEE industry is growing by more attention nowadays. Compared to virgin ABS, the price of WEEE ABS is rather cheap. Due to the problem of the degree of purity in former project, one solution of project is to investigate methods of filtering recycled ABS for FDM.

During recent years, some researches about WEEE have been carried out [11] [12].

There are similar results on their reports. With the respect to the properties of virgin ABS, a decrease in mechanical properties (except the modulus of elasticity) was recorded for both PC/ABS and ABS/HIPS mixtures [11].

One approach to different characteristics of recycled polymer was reported by Liang [13]. The contaminants of these recycled polymers are showed in Fig 11.

Previous degradation levels ABS and WEEE ABS

FTIR spectra for WEEE ABS can be compared with virgin ABS to estimate differences. According to Balart’s research, small degradation in ABS was evaluated in the FTIR spectra (Fig 12). The hydroxyl groups which absorb near 3250cm-1 shown, and the carbonyl groups, which absorb near 1750cm-1 missed in the spectra [11].

2.3.4 Mixture of ABS/PC system

According to Balart [11], Equivalent Box Model (EBM) was carried out to analysis the mechanical behavior of ABS/PC blends. EBM can obtain a range of tensile values of blends. Fig 13 shows different curves with the critical values and different A values which is an experimental parameter related to the various interaction forces along the interface.

As Balart pointed, the optimum range between10wt% and 20wt% PC in mixture shows an industrial material with balanced properties [11]. One reason is that mechanical ductile properties do not decrease [11].

Furthermore, according to Arnold, the mechanical properties of recycled ABS, especially ductility, are significantly affected by gas evolution during the process of ABS from WEEE [14].

Tarantill and Mitsakaki have done some tests about the proprieties of ABS/PC blends.

In this project, more can be dig deeper in ABS based on the Zhang’s work.

First, FTIR spectra can be used to compare the virgin ABS, after processing ABS and WEEE ABS. The relationship between WEEE ABS, after processing ABS and standard ABS should be clear in this project.

Second, ABS/PC mixture is one way to investigate. In Balart’s report, the optimum range is 10wt% to 20wt%; the mechanical properties of this range also have a good performance.

Furthermore, to refine the WEEE ABS is also one way, but due to the current situation; it is hard to find a process to increase the purity of WEEE ABS (98%).

Simply mix one equivalent part of purity ABS and one equivalent part of WEEE ABS may be possible. Higher purity of WEEE ABS may have better properties.

2.3.5 Alternative Nylon Source

According to Zhang’s research, Nylon could be an alternative material to replace the ABS P400. Compared to the ABS, Nylon have lower price. In the laboratory, nylon strimmer line has already been produced into filament with the same diameter as benchmark P400 ABS. It is easier to get the raw material. One key factor for this alternative material is melt temperature when nylon as a sample. The recommend FDM operation on standard P400 ABS was 100oC higher than the melting point. Among the nylon-6, nylon-12 and nylon-66, nylon-12, which could be a potential material, has the lowest melting point 179oC [15]. Compared to P400 benchmark, the melt mass flow rate can be a reasonable range from 7.0 to 20 [16]. During the Zhang’s trial of nylon material, 300oC is the highest temperature limitation for liquefying the FDM machine, but nylon-6 need higher. So glue gun can be used for higher FDM operation temperature. The further properties of the Nylon product need to be evaluated in the experiment.

2.4 Measuring Instruments

The target of the project is to find the alternative material to replace the ABS-P400, before that, the properties of ABS-P400 and after processing ABS should be investigated. By comparing the differences of these two materials, the possible of re-used after processing ABS is demonstrated. Moreover, the data of the standard material can be the reference of the alternative materials. The main equipments and tests may be used below.

2.4.1 Melt Flow Index (MFI)

Viscous Flow

According to Zhang’s report, viscous flow of the material is the key to this project. Low viscosity simply means that the materials are able to flow easily [2]. Viscosity is lowered significantly by increasing temperature according to the equation 4:

Where 0 is a constant, Q is the activation energy for flow, R is the universal gas constant and T is absolute temperature. In this project, MFI is used to measure the viscous flow.

According to Zhang, the some MFI results were obtained.

MFI Instrument

The Melt Flow Index (MFI) is a way to measure the ease of flow of the melt of a thermoplastic polymer. In Farlex’s free dictionary, it is defined as “the weight of polymer in grams flowing in 10 minutes through a capillary of specific diameter and length by a pressure applied via prescribed alternative gravimetric weights for alternative prescribed temperatures”[17]. One conclusion of Balart is that extrusion and injection moulding process are very important to ensure good mixing conditions and homogeneity. MFI measurements were used with an extrusion plastometer [11]. Flow rate, whose unit is g/10mins, reflects the melt polymer’s features. In this project, there are two criterions for different grades of ABS plastic. One is 2200C/10kg for injection moulding condition, which is a general standard of flow property. Another condition 2700C/2.16kg, which is a special condition for FDM condition (in Zhang’s report), need to be evaluated. Because the working atmosphere of FDM is complicated, the push force of extrusion cannot be constant due to the different speed of extrusion. To simulate the environment of FDM, the push force must be understood. The data of MFI will have the value of reference.

In this project, the difference of MFI between virgin ABS and reused ABS will be evaluated. In theory, for the thermal plastic, repeated heating and cooling cycle does not change the size of the plastic molecular chains, so MFI should not be changed [18]. However, storage conditions of reused materials are usually poor, usually placed at random, the properties of material are mostly metamorphic.

According to Zhang’s report, sample drying should be mentioned due to the absorbent of polymer. This element will greatly affect the accuracy. The specimens including ABS and nylon were put into oven for 2hours at 900C for ABS (1200C for PC) [13] and keep dry condition to prevent further water absorbing. The other factor affecting precision is cleaning. Die diameter affect the speed of melt flow. It is easy to have the deposition in the inner wall. The test result reduces due to the increasing of resistance of melt flow. So the cleaning is very important element to the accurate.

2.4.2 Capillary Rheometry Equipment

Capillary rheometer is another way to measure the viscosity of the materials. The polymer is put into a test tube with a known diameter, and there are several horizontal marks scribed on it. When put some pressure to the polymer, a vacuum is used to get the flow polymer, and the time required is measured for the melt from one mark to another. Elliott [19] explains this device also can test the density and surface tension.

2.4.3 Fourier-Transform Infrared Spectrometers

FTIR test designed in this project is to show the different molecules in the trial materials other than ABS P400. FTIR test is to evaluate what is the material composition of the FDM ABS and after FDM processing material, and to find out whether after FDM processing materials become degenerate.

Fourier-transform infrared (FTIR) is based on the interferogram, which is the idea of the interference of radiation [20]. A single is produced by the change of path length between the two beams; the changeable of distance and frequency are convertible by the mathematical of Fourier-transformation [20]. The most common interferometer used is a Michelson interferometer. There are two perpendicularly mirrors which are bisected by a semi-reflecting film (Fig 2).

The two beams defected from theses mirrors return to the beam splitter where they recombine and interfere. The moving mirror supplies an optical path difference, which controls the transmission and reflection.

2.4.4 Glue Gun

According to Zhang’s report, nylon strimmer line has been produced with the same diameter as benchmark P400 ABS. However, it was found that the strimmer line was unable to extrude from the FDM tip as need. When the temperature gets to 295 oC, which is the highest temperature limitation for liquefying the FDM machine, a small amount of material was extruded [21]. In this project, glue gun can be used to replace the tip to get higher temperature. The gun uses a heat element to continuous heat the plastic glue, the temperature of which can reach around 380 degrees [21].

2.5 mechanical rest

2.5.1 Tensile Test

The designing of tensile test aimed to evaluate the mechanical property among all alternative material.

The specimen was studied by tensile test using a Lloyd LR50KPlus materials testing machine (Fareham, UK). For each selected building material, the testing samples were produced on three patterns of raster orientation, which built up in X, Y, and Z axis respectively. All tensile test samples were made with ISO 527-2-5A standard with a dumbbell shape.

According to Zhang’s report, benchmark P400 ABS sample was test, the result shows in Fig 10.