Recycling of Polypropylene and Polyamide Blends Using Thermomechanical Recycling

INOM

EXAMENSARBETE TEKNIK,

GRUNDNIVÅ, 15 HP ,

STOCKHOLM SVERIGE 2020

Recycling of Polypropylene and

Polyamide Blends Using

Thermomechanical Recycling

Abstract

Sammanfattning

Table of Contents

1. INTRODUCTION ... 1

1.1 BACKGROUND AND ENVIRONMENTAL ASPECTS ... 1

1.2 AIM ... 2

1.3 ETHICAL ASPECTS ... 2

2. CURRENT RESEARCH ON PA6 AND PP RECYCLING ... 3

2.1 RECYCLING ... 3

2.2 COMPATIBILIZERS ... 4

2.3 3DPRINTING ... 5

2.3.1 3D printing of PA6 and PP ... 5

2.4 ANALYTICAL METHODS ... 6

3. METHOD ... 7

3.1 MATERIALS AND EQUIPMENT ... 7

3.2 PREPARATION OF THE POLYMER BLENDS ... 7

3.2.1 Extruded Filaments ... 8 3.3 EXPERIMENT ... 8 3.3.1 Shredding ... 8 3.3.2 Drying ... 8 3.3.3 Extrusion ... 8 3.3.4 3D Printing ... 8 3.3.5 Characterization ... 9 4. RESULTS ... 10 4.1 FTIR ... 10 4.2 DSC ... 13 4.3 SEM ... 17 4.4 TENSILE TESTING ... 19 4.5 3DPRINTING ... 19 5. DISCUSSION ... 21 5.1 FTIRANALYSIS ... 21 5.2 DSCANALYSIS ... 21 5.3 SEMANALYSIS ... 22

5.4 TENSILE TEST ANALYSIS ... 23

5.5 EXTRUSION AND PRINTING METHODS ... 24

6. CONCLUSION ... 26

7. FURTHER WORK/RECOMMENDATIONS ... 27

8. ACKNOWLEDGEMENT ... 28

1. Introduction

1.1 Background and Environmental Aspects

One product that uses polypropylene and polyamide-6 for the majority of the composition are carpets. After usage they mostly end up in landfills as waste, and therefore strongly

contribute to the world’s pollution. About 400000 tons are sent to landfills in the UK

annually, but this is becoming more impractical due to the raised prices for landfill dumping and less space available. One of the most common options instead is to use them for energy recovery. This is done by shredding and burning the materials. This of course comes with its own problems, since the burning process releases many toxic gases into the atmosphere and the remaining ash is dumped into landfills. The ash contains calcium carbonate which can lead to groundwater and soil pollution. Therefore, is it very important to be able to recycle these materials for future products [1].

Roughly half of the annual global solid production of plastics is thrown away each year, which is about 150 million tons which causes large amount of pollution [2]. They are mostly dumped into landfills which is harmful for the environment but also results in a huge missed market since they can be reused and recycled for new products. Recycling of plastics saves energy, roughly 130 kJ of energy can be saved by recycling one ton of plastic. This can lead to immense cost savings for our global society. One known way to recycle and reuse plastic materials is to first shred them into small granulates or other shapes appropriate for the manufacturing process which are then melted to make the new products using an extruder. The extruder will blend the melting mixture of polymers and propel it forward in the extruder using a screw [3]. This is known as thermomechanical recycling. Reusing plastic materials in this way is a good idea because the filament created can be used to create many different products through 3D-printing with complex geometries and little material waste.

The project deals with two types of polymers, Polyamides (PA6, Polyamide-66 (PA66)) and PP. PAs are characterized by their functional group CONH, and there are many varieties of this material. PA6 and PA66 are made from different monomers: Caprolactam for PA6 and a combination of hexamethylenediamine and adipic acid for PA66. PA66 are often used as replacements for metals [4]. PAs are tough materials, typically semi-crystalline with good thermal and chemical resistance [5]. PA6 fibers are used in textiles, fishing line and carpets as well as food packaging films because of their toughness and resistant properties.

PP is a linear hydrocarbon polymer and has a formula of (C3H6)n which contains a repeating

methyl group. It is a very versatile material used both as a plastic and a fiber in many different markets [6]. For example, as food packaging and consumer goods such as luggage and toys. This is also a tough and chemically resistant material, as well as an excellent electrical resistor. Other key properties are that it is translucent and has a good thermal resistance [7].

and PA6 has good mechanical properties which creates an attractive material blend, and therefore it is used for carpets.

Since the two materials PA6 and PP are critical for many applications, it is important to be able to recycle them properly. The plastic pollution is increasing and harming marine life or end up in landfills therefore recycling techniques for plastic wastes are crucial [9]. One possible solution is through thermochemical degradation which will be covered in this project.

1.2 Aim

The aim of this thesis is to recycle a model carpet waste consisting of PP and PA6 using specifically thermomechanical recycling, which is to shred the material and extrude into a filament roll. The created filament will then be tested with a 3D-printer. This recycled product will be compared to the same product made out of neat polymer materials by analyzing the structural composition after every step using various analysis techniques and testing the mechanical properties as well.

Another aspect will be to find the appropriate temperatures for the four different chambers in the extruder, since PP and PA6 have different viscosities and melting points. This will be a key part of the thesis since an even filament is needed to be able to proceed to the 3D-printing step. The temperature settings for the 3D-printer also needs to be investigated since this will affect the ability to 3D-print products. Both the bed temperature and the extruder temperature for the printer need to be calibrated.

1.3 Ethical Aspects

For this thesis, the ethical and societal aspects will not be discussed. This is because this project deals with the topic of recycling, which in general is beneficial for the society in many ways. First of all, the amount of pollution in the ocean and wildlife drastically

2. Current Research on PA6 and PP Recycling

2.1 Recycling

Thermomechanical recycling is a common recycling technique. This process is fairly straight forward. First, the plastic is washed to get rid of any debris, then shredded, melted or

extruded into filament, and remolded into the desired product or more commonly granulated into smaller granules that can be used as a type of raw material. The remolding step is usually done with the addition of some virgin material for better mechanical properties [2].

According to Mondragon G. et.al. [10], using the thermomechanical recycling for PA6 products like fishing nets should have similar main properties as the neat materials.

Therefore, it should be a viable method for new material production. Other processes include chemical recycling, which is the production of chemicals, gases, fuels or waxes with the use of catalysts. This is not as common because it is much more expensive than mechanical recycling. Another option is incineration of the materials which releases energy in the form of heat [2]. This produces ash as a biproduct and is a wasteful technique as valuable polymers cannot be reused after this process.

Carpet recycling is an aspect that needs to be improved upon. The production of carpets usually involves a lot of trimming, which results in waste products of useful polymer

material. This accounts for about 12% of the total carpet production. Recycling of carpets can be done in many ways, since there are multiple methods for recycling polymers. According to C. Mihut et. al. [11] there are four categories: primary recycling or depolymerization,

secondary, tertiary and quaternary recycling. Primary recycling is breaking down the waste product back into its original monomers, which will have equivalent quality as the neat polymer. Secondary recycling is extraction of separate polymers from the polymer blend product, not necessarily as monomers. Tertiary recycling involves melt-blending of the entire carpet blend. Quaternary recycling is energy recovery of incineration of the waste product. Tertiary recycling of carpets is a useful method, since it is very cost effective. More

specifically it consists of melting or extruding the entire carpet waste product without any polymer separation beforehand. This often results in a low-quality material since carpets consist of PA6 fibers with PP backing and latex or rubber adhesives, and these polymers are immiscible. One way of creating a recycled material with this method is to create ‘synthetic wood’. According to C. Mihut et. al. [11] the carpet waste can be shredded into lengths of around 4 centimeters, covered with prepolymers (for example epoxy resins) and cured at temperatures around 150°C-190°C and pressed into fibrous composites. These can be used for many applications, but it strongly depends of the waste carpets composition when it comes to mechanical and chemical properties for product applications. Another tertiary recycling solution is shredding and inserting the granules into a twin extruder. The melt is then pelletized and applied in injection molding. This is often used for automobile carpet waste, and this method as a result produced parts for the automobile industry for the most part. Since the use of the recycled material produced from these various methods often times is limited due to the fact that the produced material has lacking mechanical properties

2.2 Compatibilizers

When recycling plastics in general, one of the biggest obstacles is the degradation and delamination stages for mixed polymers. It is becoming more popular to use recycled

materials as feedstock. Because of this, it is very important for these materials to have useful mechanical properties, which is why compatibilizers are being used to blend different

polymers. For PP and PA plastics in the study by Gwang Ho Kim et al. [12] commingled plastic PP packaging film and PA packaging film was used. These were later separated and different blends of PA6 and PP were experimented with to test the mechanical properties. It was found that the two polymers phases are incompatible, and therefore copolymers have to be used as compatibilizers to obtain improved mechanical properties. This was only achieved with the addition of copolymers with maleic anhydride (MA). For example, PP-g-MA is one of the compatibilizers that showed a great improvement in the mechanical properties, and thus indicating that these copolymers are beneficial for the recycled material blends from waste plastics.

There are different types of compatibilizers that could be used, but for example in the study by S. S. Dagli et al. [9] an acrylic acid grafted modified PP was used (PP-g-AA) to combine the two phases when melt reprocessing these two plastics. The study showed that

compatibilizer effects the morphology and the blends properties significantly in a positive way.

In the study by F.P. La Mantia [13] the recycling ability of compatibilized and

uncompatibilized PA6/PP blends were researched, and the materials that were compared were 4 times recycled in an extruder and virgin polymer blends containing PA6, PP, (maleated polypropylene) PPMA and poly(propylcrylic acid) (PPAA). This study showed that the recycled and the virgin blends showed no observable changes in the morphology. The reprocessing created copolymers PA-g-PP which slightly improved the morphology, meaning that the adhesion of the dominating polymers in the mix (PP and PA6) improved after the processing extrusions, but the molecular weight decreases. It was also discovered that the mixing time for the reprocessed polymers is crucial, and that with longer times the adhesion worsens and with that the mechanical properties. Therefore, extrusion is a more useful technique here since it is milder than for example degrading in a mixer. This is true for the uncompatibilized blends and the blends compatibilized with PP functionalized with acrylic acids, but no changes were noted for the compatibilized blends with maleic acid. In

conclusion, the extrusion recycling technique does not seem to worsen the polymer blends mechanical properties. Also, according to F.P. La Mantia[13] the viscosity of the blends with compatibilizer increases compared to neat PP/PA6 blends, meaning that the compatibility of the two phases is much better.

For products such as carpets, the most common ways to disposed of them is to either burn them for energy return or they end up in landfills. Only 2% of all carpets that end up as waste go through fiber reprocessing. This involves the depolymerization which leads to the PA fibers to separate from the rest of the material. This process results in the recovery of

2.3 3D Printing

3D printing is one of the fastest growing technologies for product manufacturing. This is because it uses less material to produce products with complicated geometries, resulting in a more efficient and cost-effective way to produce parts. 3D printing is a method used to produce 3D objects with controlled layer deposition of a printable material of choice to produce the final structure using CAD models [14]. Many different polymer materials can be used for 3D printing, including acrylonitrile butadiene styrene (ABS), polylactic acid (PLA), polycarbonate (PC), and of course polymers present in carpets such as PA6 and PP. The products made from these materials using this method can be applied in many industries such as aerospace, architecture, structural models, art fields, medical fields and many others. But the structures made from pure polymers are usually only used at prototypes due to their lacking mechanical properties. Therefore, composite products are favored, with fiber or particle reinforcement [15].

The most common 3D printing method is fuel deposition modelling (FDM). These printers work by extruding filament. This is done by melting the filament into a semi-liquid state at the nozzle, which is then extruded layer by layer to compose the desired structure using computer software instructions. The layers solidify and fuse together to form the solid product [14,15]. According to Wang X. et. al. [15] reinforcing the 3D printer filament with fiber or particles has shown to improve the mechanical properties significantly. When this is done, it can be seen that the tensile strength of PA reinforced materials produced using FDM can increase up to 446%, depending on the amount of reinforcement used. This is interesting since with fiber-based reinforcement, there are more voids present due to the poor bonding between the matrix and reinforcement, the material has similar strength to compression molded pure polymer materials that contain nearly no voids. This is due to the fact that there are many more fibers aligned in the load-bearing direction.

2.3.1 3D printing of PA6 and PP

3D printing with PA6 and PP by themselves can be challenging, as well as using a

2.4 Analytical Methods

The techniques used for evaluating the chemical composition of the materials tested in this thesis will be DSC, FTIR and SEM, and mechanical properties with tensile testing.

DSC (Differential scanning calorimetry) is an analysis technique used to detect how a material’s heat capacity is changed with increasing temperature. This heat analysis is used to measure the melting points of the different compounds in material blends [18]. The results can then be compared to the melting points of the materials that should be present to confirm if this is the case [19].

FTIR (Fourier Transform Infrared Spectroscopy) is a method to analyse the composition and structure of different molecular mixtures by exposing the material to infrared light which is absorbed at specific frequencies depending on the functional groups present [20]. With this the composition of the material that will be recycled can be compared to raw PP and PA. Scanning electron microscopy (SEM) uses a bean of electrons that are propelled towards a sample and results in an image of the surface. The image can contain information about morphology (texture), and the orientation of the different materials present in the sample, depending on the settings. This is obtained due to the electron-sample interactions, and the magnification of this analysis varies from 20X to roughly 30,000X [21].

3. Method

The neat and recycled filament were made using the mechanical recycling method. The neat filament and products were produced first to set up the extruder with the correct settings for the different chambers and tested in the 3D-printer. Afterwards the carpet was processed in the same manner. The chemical composition was evaluated after each step, and the

mechanical properties were tested using a tensile test.

3.1 Materials and Equipment

Neat PP and PA6 were received from Domolen and Akulon respectively. The material used for recycling is a common carpet that contains PP and PA6 for the majority of its

composition. The compatibilizer PP-g-MA is from the company Sigma Aldrich. Below is

Figure 1with an image of the repeating structures for the different polymers used in this thesis.

Figure 1: The structures of the three materials used – PP (top), PA6 (middle) and PP-g-MA (bottom).

The equipment used are a shredder, extruder and a 3D-printer. The shredder and extruder are from the company 3Devo. The 3D printer is produced by CreatBot and is the model DX_1. To be able to use the 3D-printer, a model has to be made using programs that create .gcode files, for this the programs Ultimaker Cura 4.5 and CreatWare were used.

3.2 Preparation of the polymer blends

3.2.1 Extruded Filaments

• 1x extruded PA6/PP + 5% Compatibilizer • 1x extruded PA6/PP

• 2x extruded PA6/PP

• 2x extruded PA6/PP + Carpet

• 1x extruded PA6/PP + 5% Comp + 5% (Carpet + 5% Compatibilizer) • 1x extruded PA6/PP + Carpet without backside

For the material with carpet without the backside, it is meant that the rubber layer on the bottom of the carpet was removed to achieve an even filament without rubber pieces and only containing PA6 and PP.

3.3 Experiment

3.3.1 Shredding

The shredding process involved breaking down the polymer pellets and the carpet pieces into smaller grains, that would result in better blending and a material with better mechanical properties. The process is fairly straight forward, the 3Devo shredder was used and

granulated the material inserted into smaller pieces which were then processed further with the next methods.

3.3.2 Drying

The shredded material was then placed into a vacuum oven at room temperature. This step is necessary due to the fact that the polymers (mostly PA) absorbs moisture from the

surroundings quite easily. This results in uneven filament when extruding due to the fact that air bubbles are created due to the high heat in the extruder. The uneven filament then cannot be used further in a 3D-Printer since the material does not meet the requirements for the printing process. The shredded material was dried in vacuum for at least 8 hours before use.

3.3.3 Extrusion

For this process the challenge was finding the correct temperature for the 4 different chambers in the extruder. After trial and error and literature study, the settings used for created PA6/PP filament was: 180-210-235-240°C. 180°C is for the chamber furthest from the extrusion point, meaning nearest to the hopper. 240°C was near the extrusion point, where it is very important that the material is liquid and can pass through the extraction point to not congest the extruder. The speed used was 4.0 rpm and a fan speed varying from 80-100%. Since the filament is meant for the 3D printer a diameter for it was set at around 1.65mm to not exceed the limit of 1.75mm for the printer.

3.3.4 3D Printing

was provided with the printer for better adhesion. The structure that was printed was a dog bone for tensile testing.

3.3.5 Characterization

To be able to evaluate the different chemical changes and the polymers present, analytical test were performed such as DSC, FTIR, tensile test and SEM. This was done for various filaments during different stages of the dog bone product development. Meaning after shredding, extrusion and 3D printing.

3.3.5.1 DSC

To perform the DSC analysis first the sample had to be weighed to be between 5-10g and placed in an aluminum cup which was then placed into the machine. The apparatus goes through 5 different stages of heating, cooling and isotherms to set all of the samples to the same starting reference point. The curve that was used for analysis was the heating, stage 5 which was from -30°C to 300°C. To compare the results, the values for melting points and glass transition temperatures are used from PerkinElmer [19], shown below in Table 1:

Table 1: Melting points and glass transition temperatures from PerkinElmer for reference

Polymer Melting Point (Tm) [°C] Glass Transition (Tg) [°C]

PA6 210-220 40-60

PP 165-175 (-20) – (-5)

3.3.5.2 FTIR

The apparatus used was PerkinElmer spectrum 100 FT-IR-Spectrometer with the sample holder Specac golden Gate. The software used was PerkinElmer Spectrum. The graphs produced were transmission spectrums. First, the program was run without any sample to remove the effect of the surroundings, and afterward the necessary samples were tested. The values for the spectrum analysis of the different functional groups and bonds present are used from ResearchGate and Sigmaaldrich figures and tables [23,24,25].

3.3.5.3 SEM

4. Results

4.1 FTIR

FTIR results were grouped based on the type of comparison that was necessary for the thesis. The references of the neat polymers are present in figure 2 as well as an evaluation of the different parts of the carpet in figure 3. The remaining figures, figures 4-7, are the results from the analysis of the extruded filament, with different compositions of carpet and compatibilizer for comparison.

Figure 2: References for the polymers used with the indicated vibrations detected for the different polymers

PA 6 PP PP-g-MA

Name

Sample 006 By Administrator Date Thursday, February 06 2020 Sample 005 By Administrator Date Thursday, February 06 2020 Sample 011 By Administrator Date Thursday, February 06 2020

Figure 3: All the parts of the carpet used

Figure 4: 65% PA6 and 35% PP Polymer blends comparison without compatibilizer, extruded either 1 or 2 times

Carpet upper part, small pieces carpet weave (middle layer) carpet fabric (top side) Carpet backside (rubber)

Name

Sample 008 By Administrator Date Thursday, February 06 2020 Sample 010 By Administrator Date Thursday, February 06 2020 Sample 007 By Administrator Date Thursday, February 06 2020 Sample 009 By Administrator Date Thursday, February 06 2020

Description 4000 3500 3000 2500 2000 1500 1000 600 101 39 40 45 50 55 60 65 70 75 80 85 90 95 cm-1 %T PA6PP PA6PP x1 PA6PP x2 Name

Sample 004 By Administrator Date Friday, February 28 2020 Sample 005 By Administrator Date Monday, April 06 2020 Sample 003 By Administrator Date Monday, April 06 2020

Figure 5: PA6/PP polymer blends comparison between extrusion twice and one time and with/without carpet

Figure 6: Effect of PP-g-MA compatibilizer on polymer blends

PA6PP x2 + 5% carpet PA6PP x2

PA6PP

PA6PP x1 + 5% carpet backside removed Name

Sample 002 By Administrator Date Monday, April 06 2020 Sample 003 By Administrator Date Monday, April 06 2020 Sample 004 By Administrator Date Friday, February 28 2020 Sample 000 By Administrator Date Monday, April 06 2020

Description 4000 3500 3000 2500 2000 1500 1000 600 101 62 65 70 75 80 85 90 95 100 cm-1 %T PA6PP x1 + 5 % comp

PA6PP x1 + 5 % comp + 5% carpet with 5 % comp PA6PP 5% PP-g-MA Extruded twice

PA6PP x1

Name

Sample 000 By Administrator Date Monday, April 06 2020 Sample 001 By Administrator Date Monday, April 06 2020 Sample 003 By Administrator Date Friday, February 28 2020 Sample 005 By Administrator Date Monday, April 06 2020

Figure 7: Comparison of 3D printed PA6/PP blend to extruded filament

4.2 DSC

The DSC analysis was first performed for the reference polymers which are present in figures 8-10. Also, the carpet has been analyzed and the rubber backside by itself to show that the composition of the carpet is much different from the polymers being recycled. This can be seen in figures 11-12. 6 different polymer blends with and without recycled carpet and compatibilizer have been analyzed as well and compared with the reference peaks from figures 8-10. These images can be seen in figures 13-18.

Figure 8: DSC analysis of PP, with the value of the melting peak at 160.27°C

PA6PP print +5 % carpet without backside x1 extruded PA6PP print x1

PA6PP

Name

Sample 000 By Administrator Date Wednesday, April 22 2020 Sample 001 By Administrator Date Wednesday, April 22 2020 Sample 004 By Administrator Date Friday, February 28 2020

Figure 9: DSC analysis of PA-6, with the value of the melting peak at 220.62°C

Figure 10: DSC analysis of PP-g-MA polymer, two peaks were noted

Figure 11: Carpet components analysis, two melting peaks were noted

Figure 12: Analysis of only the backside of the carpet, one peak noted

Figure 13: DSC analysis of PA6/PP blend, multiple peaks including a glass transition

Figure 14: DSC analysis of PA6/PP blend with compatibilizer

-30 20 70 120 170 220 270

Temp [°C]

Carpet, backside only

Peak: 215.97 °C -30 20 70 120 170 220 270 Temp [°C]

PA6/PP Blend

65%PA6/35%PP + 5% PP-g-MA, single

extrusion

Peak: 159.68 °C

Peak: 219.44 °C

Figure 15: DSC analysis of PA6/PP polymer blend with compatibilizer, and extruded twice

Figure 16: DSC analysis of PA6/PP blend with 5% carpet without backside

Figure 17: DSC analysis of PA6/PP blend 3D printed

-30 20 70 120 170 220 270

Temp [°C]

65%PA6/35%PP + 5% PP-g-MA, extruded

twice

65%PA6/35%PP + 5% carpet without backside

Figure 18: DSC analysis of PA6/PP blend with 5% carpet without backside, and 3D printed

4.3 SEM

The images from the SEM analysis can be seen in figures 19-25.

Figure 19: SEM visuals of PA6/PP polymer blend extruded 1x fracture surface, with zoom (from left to right) 60X, 100X

Figure 20: SEM visuals of PA6/PP polymer blend with 5% compatibilizer extruded 2x fracture surface, with zoom (from left to right) x60, x100, x500

-30 20 70 120 170 220 270

Temp [°C]

65%PA6/35%PP + 5% Carpet without backside

3D-Printed

Peak: 212.87°C Peak: 160.42°C

Figure 21: SEM visuals of PA6/PP polymer blend with 5% carpet without the backside extruded 1x fracture surface, with zoom (from left to right) x250, x500, x2.5k, x5.0k

Figure 22: SEM visuals of PA6/PP polymer blend with 5% carpet without the backside 3D printed 1x fracture surface, zoom x500

Figure 23: SEM visuals of different parts of the PA6/PP polymer blend with 5% carpet with backside removed and 3D printed 1x fracture surface, zoom x5.00k

Figure 25: SEM visuals of PA6/PP polymer 3D printed 1x fracture surface, zoom (from left to right) x500, 2.50k, x5.00k, x5.00k

4.4 Tensile Testing

The tensile test was performed with each type of filament tested 5 times. The 3 values that were recorded are modulus, tensile strain and tensile stress. 5 different types of filament were tested, and the tables show the values including the standard deviation for each test. This can be seen in Figure 26

Figure 26: the three measured parameters during tensile testing with 5 different filament types. Green – Tensile Stress, Blue – Modulus, Orange – Tensile Strain (Extension)

4.5 3D Printing

5. Discussion

5.1 FTIR Analysis

From the reference figure 2 the functional groups transmission spectra points are shown. Comparing this to the carpet material in figure 3, it is clear that the carpet contains the correct polymers compared to the reference in figure 2, since the spectrums are very similar and have similar peaks. The reference peaks were found using a table of IR Sprectum values from SigmaAldrich and from 2 figures from Researchgate [23,24,25].

For figures 4-7, there is an orange box present around 3300 cm-1 _{to highlight the transmission}

spectrum at this point since there is a drastic difference in peak height compared to the reference in figure 2. The peak around 3300 cm-1 _{references to the N-H bond present in}

polyamides, in this case PA6 is much smaller in figure 4-7, compared to figure 2. The filaments shown in these figures have been subjected to the thermomechanical recycling process, and this could then of course lead to polymer degradation. In any case, the PA-6 is still present since the peak around 3300 cm-1 _{indicating the N-H is still present and the other}

amide peaks are visible, but this shows that there is some degradation of the chemical composition and the chain structure of the material after processing. According to M.P. La Matina [13] the longer the mixing time (shredding) of the PP/PA-6 mixture is, the more degradation of the polymers is observed due to the breakage of some the PP/PA6 bonds that have been present, therefore the materials shredded more than once show a smaller

transmission peak as seen in figure 5 where the filament that has been extruded twice has much fewer and smaller peaks.

For the majority of the filaments tested, the same chemical composition is present at all times, but it can be noticed that with more processing steps for most of the filaments there is more degradation present, but mostly for the amide indicator peaks specifically the N-H and C-N spectrum indicators. This is because the other peaks present represent the different types of C-H bonds present in the filament’s polymers. Both PA6 and PP have many of these bonds, meaning that there are many more present in the polymer blends tested and will therefore absorb the infrared light and result in peaks.

One observation is that for the materials analyzed after 3D printing, figure 7, the peaks seem to be larger and more prominent in these samples even though they have been processed more. But this could because only a small fraction of the filament is tested, and so a full evaluation of the entire extruded and 3D printed material cannot be evaluated and compared.

5.2 DSC Analysis

The results from the DSC analysis show that the compounds present in the different filaments do not change dramatically after the different processing stages. Figures 8-11 are used as references to compare if any drastic changes in the polymer blends occur after processing. The values from PerkinElmer will be used to evaluate the results (Figure 1). What is seen is that the values for Figures 8-11 are similar to the reference from PerkinElmer, meaning that the polymers used are correct.

Figure 12 is an analysis of the backside of the carpet, which is presumably rubber since it has physical similarities to rubber compounds. Because of this the backside of the carpet could not melt, but there is still a peak at around 216°C, which coincides with the melting point of PA6. The reason behind this could be that the backside was not properly removed from the PA6 fabrics, since it is very hard to determine without magnification on the object if the rubber has been cleanly separated.

Figure 13 shows a polymer blend of PA6 and PP. What is noticed is that the melting points and glass transition temperature reflect that of those polymers, but lower than that of the references that where analyzed separately in figures 8 and 9. This could be due to the shredding process which grinded the polymer pellets into smaller pieces, which could have then broken down the polymer chains present. The shorter the chain length the lower the melting point will be for the polymer [26]. This in itself is a sign of degradation of the polymer material after one processing step for reuse of the material.

The effect of shredding and reprocessing polymers can be detrimental to some of their mechanical properties, due to the fact that the chains are broken down and degradation occurs. This can be seen when comparing figure 14 and 15 which both contain a

compatibilizer but have been extruded different amount of times. The filament from figure 15 has gone through the thermomechanical process twice (shredding and extruding). The DSC image of figure 15 does not resemble figure 14 whatsoever. Clearly the thermomechanical process has affected the polymer chains and broken them down, and it is noticed that the PA6 melting peak has disappeared. The curve was increasing towards the end of the heating cycle, but since the machine did not reach higher than 300°C no peak is observed.

The other filaments did not have much variation from the references, the melting points and glass transition temperatures were roughly that of the starting material except for the 3D-Printed filament with 65% PA6 and 35% PP with 5% carpet of the total weight. Here it is observed that there is a melting point at around 130°C which does not correspond to any of the starting materials. A possible explanation for this could be that the carpet or the polymer blend contained undocumented debris. According to the PerkinElmer values, this peak could be polyethylene (PE) which is a very common polymer but should not be present in the carpet according to its specifications. High-density polyethylene (HDPE) was used to clean the extruder after the process, and so the tested sample could have contained some HDPE since it does have a similar melting point of around 130°C.

5.3 SEM Analysis

The SEM images help evaluate and understand the chemical composition of the material, and how the structure inside of the material look like. When comparing the images from figure 19 and 20, which are filaments extruded once without and with the compatibilizer PP-g-MA. Figure 19, which is without the compatibilizer, the fracture surface has many more pulled out fibers. This is due to the fact that without PP-g-MA the two polymers PP and PA-6 are immiscible. A possible explanation for this is that since the two polymers are not compatible. The immiscibility leads to the PA6 fibers being more dispersed and longer, resulting in them taking up most of the tension during tensile testing. Compared to previous studies that have been mentioned in the literature study, such as the study by Gwang Ho Kim et.al. [12] that stats that the compatibilizers effect is that there is less dispersed phase and increased

An interesting observation is that in figure 21 and 22 the addition of 5% carpet impacted the composition drastically. Here it is noticeable that the filament is built up with what seems to be layers, unlike the structure of the blends without carpet, figure 19 and 21. This could result in better mechanical properties in the direction of the layers. The PA6 fibers look to be very fine.

Since 3D-Printing consists of an extrusion process, the filament is therefore processed once more before the final product is developed. The filament extruded from the 3D-printer onto the surface, is much thinner than the starting material. This in itself theoretically should lead to a finer composition with shorter polymer chains, resulting in a denser and more compact product which hopefully results in better mechanical properties. What can be noticed in figure 23 and 24 in comparison to the filament with carpet but that has not been printed, the 3D printed composition seems to be much finer if comparing in the same magnification. A possible reason for these changes could be explained with the help of the DSC and FTIR analysis. As was previously seen, the chemical composition changes with more processing steps. For example, the filament that was extruded twice showed no sign of a melting point within the limits for PA6. This could mean that the 65% content of the PA6 fibers have been degraded by the recycling processed, and therefore are almost unnoticed in the 3D printed SEM images.

When printing with pure PA6/PP polymer blend without the addition of recycled carpet, what is observed in the SEM images in figure 25 is that there seem to be pores in the material. This could be due to the printing process, since the printer was set to very high temperatures and the material did not print as well as expected. The printer setting could not be set up correctly for this printing process even after extensive literature study and trial and error.

5.4 Tensile Test Analysis

With the addition of tertiary recycled material, through thermomechanical recycling

according to Mihut C. et.al. [11] these materials often times are used for low quality products. This must be because the material retrieved from this type does not have the best mechanical properties. What is noticed from the tensile results collected during this thesis, this does not seem to be the case if small amounts are added. The tensile test results with carpet are similar to the ones without recycled carpet, and even better. Only 5% of the recycled carpet was added and so no general conclusion as to how a large amount of recycled carpet would affect the mechanical properties.

The tensile testing could not be performed on the printed dog bones, since only two were made and so no conclusive data could be retrieved from such tests. But, according to Wang X. et. al. [15]reinforcing the 3D printer filament with fiber or particles has shown to improve the mechanical properties significantly. This could be true also for extruded material, since a 3D printer consists of an extruder just in smaller fashion. The tensile test results seem to agree with this statement, since the material with the carpet has the higher Youngs modulus, this could be due to the rubber particles present.

immiscible phases blend much better as seen in figure 20 and this then leads to fewer PA6 long fibers that take up the majority of the tension. What is observed is that the

compatibilizer does not seem to have a positive effect on the mechanical properties like the literature said, but more the added recycled carpet. This could be because the compatibilizer has similar functional groups to PP, and so when broken down it could result in better compatibilization and also more PP in the material which has worse mechanical properties than PA6. But compared to the study Gwang Ho Kim et.al. [12] which states that the compatibilizers effect is that there is less dispersed phase and increased interfacial adhesion which in theory should lead to better mechanical properties if the morphology is more

packed. The difference between that study and this thesis, is that different amounts of PP and PA6 (75%PP and 25%PA6) were used. This can affect the final products properties

compared to what was produced during this thesis.

According to Mondragon G. et.al. [10], using the thermomechanical recycling process for specifically PA6 waste like fishing nets should result have similar main properties as the neat materials that have not been processed. From the 5 tests performed for the 5 different

materials in this thesis, this does not seem to coincide with this study’s findings. The mechanical properties are affected significantly. This could be because only 5 tensile tests were performed and so the standard deviation is very large, and the results are not conclusive but provide an overview.

5.5 Extrusion and Printing Methods

The process for recycling during the course of the thesis had been altered multiple times. This was to try and produce the best filament possible for 3D printing, which was the goal. The shredding process was straight forward and did not need changing. When the material was not used after the shredding process, it was placed into the vacuum so that moisture would not be absorbed.

better morphology as well. Another reason could be that the settings for these two materials needed to be much different, but this does not seem to be the case since the settings were altered multiple times according to various literature and nothing improved the printability of the neat blend.

A lower temperature for the printing was tested but did not result in a better constructed product. Any lower temperatures could not have been tested since this was below the

6. Conclusion

To conclude, using thermomechanical recycling for carpet waste and 3D printing with the filament is a viable recycling option. From these analysis techniques it is also noticed that some degradation of the polymers occurs. According to the DSC and FTIR analysis it is more likely that the PA6 degrades at a higher rate than the PP, since the amide indicators of

spectrum peaks and melting points are smaller or even not existing on some of the results. When looking at the SEM images, it is clear that there is a composition change with the addition of the compatibilizer PP-g-MA since the PA6 fibers seem to be fewer. OF course, these tests are conducted for a limited amount of the material filament, and so no conclusive statement can be made for the whole filaments.

The mechanical properties seem to be rather similar with and without the addition of recycled carpet when comparing the Youngs Modulus, but the extension and tensile stress are worse than for neat PP/PA6 blends. The PA6 fibers take up most of the tension in the neat samples, and therefore it is difficult to compare to a blend with better compatibility.

Recycling carpets using thermomechanical recycling can definitely be a possibility and will result in similar properties compared to neat polymers, but this is only for small additions of carpet. How this can be implicated on an industrial level is difficult to analyze since the removal of the rubber backside from the carpet was in itself a challenging and

7. Further Work/Recommendations

What is recommended based on this report, is that more research is needed for large recycling attempts with carpet waste. Configuring the 3D printer with the right setting for the specific PA6/PP blend is very difficult, and an adhesive surface needs to be used and many more trial and error attempts are in order.

More tensile tests should be performed to conclude the carpets effect on the mechanical properties and as to why the neat blend has great mechanical properties. Many more tests should be performed for the different analysis methods to conclude the degradation effects of the 3D printing and thermomechanical recycling process to be able to understand the

mechanical properties better.

8. Acknowledgement

First and foremost, thank you to my supervisor and doctoral student Eva Bäckström from the Polymer technology division at KTH who has helped me tremendously during this thesis and guided me in the right directions when needed. Thank you for being at the lab with me and helping me finish the laboratory work during these tough pandemic times. Also, thank you to professors Karin Odelius and Minna Hakkarainen for helping to set up this thesis and

9. References

[1] A. Sotayo, S. Green, and G. Turvey, “Carpet recycling: A review of recycled carpets for structural

composites,” Environmental Technology & Innovation, vol. 3, pp. 97–107, 2015, doi: 10.1016/j.eti.2015.02.004 [2] J. M. Garcia and M. L. Robertson, “The future of plastics recycling,” Science (New York, N.Y.), vol. 358, no. 6365, pp. 870–872, 2017, doi: 10.1126/science.aaq0324

[3] Grigore, M. E. (2017). Methods of recycling, properties and applications of recycled thermoplastic polymers. MDPI, 2(4), 1–11. https://doi.org/10.3390/recycling2040024

[4] "Polyamide (PA) or Nylon: Complete Guide (PA6, PA66, PA11, PA12…)", Omnexus.specialchem.com. [Online]. Available: https://omnexus.specialchem.com/selection-guide/polyamide-pa-nylon . [Accessed: 10- May- 2020].

[5]"Nylons (Polyamide)", Bpf.co.uk, 2020. [Online]. Available:

https://www.bpf.co.uk/plastipedia/polymers/polyamides.aspx . [Accessed: 15- Mar- 2020].

[6]"Polypropylene (PP)", Bpf.co.uk, 2020. [Online]. Available:

https://www.bpf.co.uk/plastipedia/polymers/PP.aspx#properties . [Accessed: 15- Mar- 2020].

[7]"Polypropylene (PP) Plastic: Types, Properties, Uses & Structure Info", Omnexus.specialchem.com. [Online].

https://omnexus.specialchem.com/selection-guide/polypropylene-pp-plastic [Accessed: 01- May- 2020].

[8]Pal, S. K., & Kale, D. D. (2000). Effect of processing conditions and properties of PP/nylon 6 blends. Journal of Polymer Research. https://doi.org/10.1007/s10965-006-0110-3

[9] S.S. Dagli, M. Xanthos, J. A. B. (1992). Blends of Nylon 6 and Polypropylene with Potential Applications in Recycling. In Emerging Technologies in Plastics Recycling (pp. 241–257). American Chemical Society. [10]G. Mondragon, G. Kortaberria, E. Mendiburu, N. González, A. Arbelaiz, and C. Peña‐Rodriguez, “Thermomechanical recycling of polyamide 6 from fishing nets waste,” Journal of Applied Polymer Science, vol. 137, no. 10, p. n/a–n/a, 2020, doi: 10.1002/app.48442

[11] Mihut, C.; Captain, D. K.; Gadala-Maria, F.; Amiridis, M. D. Review: Recycling of nylon from carpet waste. Polym. Eng. Sci. 2001.

[12] Gwang Ho Kim, Seung Sang Hwang, Bong Gyoo Cho, S. M. H. (2007). Reactive Extrusion of Polypropylene and Nylon Blends from Commingled Plastic Wastes. Wiley InterScience.

[13] F.P. La Mantia, L. C. (2000). Recycling of compatibilized and uncompatibilized nylon/polypropylene blends. Elsevier Science Ltd.

[14] Sadasivuni, K.K., Deshmukh, K. & Al-Maadeed, K.K., 2020. 3d and 4d printing of polymer nanocomposite materials: processes, applications, and challenges.

[15] X. Wang, M. Jiang, Z. Zhou, J. Gou, and D. Hui, “3D printing of polymer matrix composites: A review and prospective,” Composites Part B, vol. 110, pp. 442–458, 2017, doi: 10.1016/j.compositesb.2016.11.034

[16]"Ultimate Materials Guide - 3D Printing with Nylon", Simplify3d.com. [Online]. Available:

https://www.simplify3d.com/support/materials-guide/nylon/ . [Accessed: 12- Apr- 2020].

[17]"Spectrum Filaments", Spectrumfilaments.com, 2019. [Online]. Available:

https://spectrumfilaments.com/data/include/cms/Spectrum_Catalogue_2019.pdf . [Accessed: 13- Apr- 2020].

[18] Menczel, J. D., Judovits, L., Prime, R. B., Bair, H. E., Reading, M., & Swier, S. (2008). Differential Scanning Calorimetry (DSC). Thermal Analysis of Polymers: Fundamentals and Applications, 7–239.

[19] "MP, TG, an Structure of Common Polymers", Perkinelmer.com. [Online]. Available:

https://www.perkinelmer.com/CMSResources/Images/44-74863TCH_MPTGAndStructureOfCommonPolymers.pdf . [Accessed: 20- Mar- 2020].

[20]"FTIR Spectroscopy", Mt.com. [Online]. Available:

https://www.mt.com/int/en/home/products/L1_AutochemProducts/ReactIR/ftir-spectroscopy.html?cmp=sea_01010123&SE=GOOGLE&Campaign=MT_AC_EN_ROW&Adgroup=In+Situ+ Analysis++FTIR++Exact&bookedkeyword=ftir%20analysis&matchtype=e&adtext=403854681266&placement =&network=g&kclid=_k_EAIaIQobChMIqtPqs8Hs5wIVhswYCh1lGQxREAAYAiAAEgKlwvD_BwE_k_&gc

lid=EAIaIQobChMIqtPqs8Hs5wIVhswYCh1lGQxREAAYAiAAEgKlwvD_BwE . [Accessed: 16- Apr- 2020].

[21]"Scanning Electron Microscopy (SEM)", Geochemical Instrumentation and Analysis. [Online]. Available:

https://serc.carleton.edu/research_education/geochemsheets/techniques/SEM.html . [Accessed: 18- Apr- 2020].

[22] M. Jawaid, M. Thariq, and N. Saba, Mechanical and physical testing of biocomposites, fibre-reinforced composites and hybrid composites. 2019.

[22] L. A. Díaz-Alejo, E. C. Menchaca-Campos, J. Uruchurtu Chavarín, R. Sosa-Fonseca, and M. A. García-Sánchez, “Effects of the Addition of - and -NH Substituted Tetraphenylporphyrins on the Structure of Nylon 66,” International Journal of Polymer Science, vol. 2013, no. 2013, pp. 1–14, 2013, doi: 10.1155/2013/323854 [23]"IR Spectrum Table & Chart", SigmaAldrich. [Online]. Available:

https://www.sigmaaldrich.com/technical-documents/articles/biology/ir-spectrum-table.html#table-by-group . [Accessed: 12- May- 2020].

[24] N. Baig and T. A. Saleh, “Superhydrophobic Polypropylene Functionalized with Nanoparticles for Efficient Fast Static and Dynamic Separation of Spilled Oil from Water,” Global Challenges, vol. 3, no. 8, p. n/a–n/a, 2019, doi: 10.1002/gch2.201800115.

[26]"Changing the properties of polymers and plastics", Rsc.org, 2020. [Online]. Available:

https://www.rsc.org/Education/Teachers/Resources/Inspirational/resources/3.1.5.pdf . [Accessed: 01- May-