*2.3. Characterization*

The CFRPs wastes are characterized using Thermogravimetric Analysis (TGA), Scanning Electron Microscopy images and Energy-dispersive X-ray spectroscopy (SEM-EDS). Netzsch instrument TG 209 F3 model is used to make TGA measurements. The measurements were characterized for the thermal degradation curve of the CFRP characterized under an air atmosphere. The weight of the sample is 10 mg. Each sample was heated from room temperature to 650 ◦C at a heating rate of 10 ◦C/min. Three target areas were identified for each of the rCF samples, and taking the core layer of each sample, the surface element (C, O, and Si) of the core layer is measured using a Hitachi TM 4000 Plus instrument with Oxford AZtecOne (EDS) to assess removal capability and uniformity for silane and resin. The high magnification physical morphology of carbon fiber was detected by Hitachi Regulus 8100 field-emission scanning electron microscopes (FE-SEM).

**Figure 1.** CFRPs waste recycling treatment by micro-wave pyrolysis system.


**Table 1.** Details of samples at different reaction conditions.

\* There was no retention time at the desired temperature.

#### **3. Experimental Results and Discussions**

#### *3.1. TGA Analysis of CFRPs Waste*

It is important to understand the thermal decomposition curve of the CFRPs' waste from bicycle frame waste, since cause each batch of CFRP could be from different source. As shown in Figure 2, the DTG curve has a significant mass loss in the range of 250 to 450 ◦C and 490 to 560 ◦C, respectively. Therefore, it can be defined that the decomposition temperature of the first TGA stage is 220 to 450 ◦C and the second stage is 475 to 550 ◦C. The third stage of the decomposition range is between 600 and 650 ◦C. The thermal decomposition curve is consistent with results presented by Yatim et al. [57] and Deng et al. [55].

In Figure 2B–D, the SEM image also shows the extent to which CFRP waste is removed from the resin between the fibers of the CFRPs' waste at various stages of TGA. In stage 1, at 330 ◦C (Figure 2B), a weight loss of approximately 5% occurred. It can be observed that there is still a large amount of epoxy residue between the fibers. In stage 2 (Figure 2C), after heating beyond the first interval, the weight loss is 30%. The gaps between the fibers can be clearly seen and there is still un-removeable resin around the carbon fiber. After 500 to 600 ◦C, second interval, an increasing amount of resin is removed; approximately 20%. No residual resin was observed on the surface of carbon fiber. Based on TGA thermal weight loss results, it can be speculated that the carbon fiber-to-resin weight ratio of the CFRPs waste is approximately 40:60.

**Figure 2.** TGA analysis of CFRP waste: (**A**) CFRP of thermal decomposition curve; and (**B**–**D**) SEM micrograph of CRFP thermal decomposition stages (**B**) stage 1, (**C**) stage 2, and (**D**) stage 3.

#### *3.2. Compare the Pyrolysis Result of Microwave with Traditional Treatment for CFRPs Waste*

The parameters obtained from TGA thermal decomposition were used as the reference condition for the removal of the resin by high temperature furnace and microwave pyrolysis. Figure 3 presents CFRP waste samples recovered using two different heat treatment methods. Figure 3A shows SEM image and EDS analysis results of virgin waste carbon fiber. The SEM image clearly shows a significant quantity of resin wrapped around the carbon fiber. EDS elemental analysis shows the presence of carbon, oxygen, and silicon content at 85.8 ± 5.3%, 13.6 ± 4.7%, and 0.3 ± 0.6%, respectively. Figure 3B shows no significant difference between CFRP waste in high-temperature furnaces after heating up to 650 ◦C, except for a color change resulting in a darker appearance. And in Figure 3C, CFRP waste decomposed by microwave pyrolysis results in carbon fibers that can be separated easily into individual fibers.

#### *3.3. Influence of Different Temperature on Removed Resin in Large Scale Waste CFRP for Microwave Pyrolysis*

In this study, the samples of waste CFRPs were obtained from recycled bicycle frames. Samples were prepared to the required sizes through smash and cut methods; the thickness being approximately 4.5 mm. The CFRP was closely composed of multilayer CF and resin. The pyrolysis of the resin requires enough oxygen and heat for carbon oxidation to occur and for effective removal of resin from CFRP wastes. The irregular and curled shape of CFRP wastes and the compact multilayer CF structure could easily result in non-uniform hot spots in the inside of each waste element and this structure also makes it difficult for the resin to make good contact with surrounding air. Therefore, the resin removal methodology requires further research [56].

To understand the uniformity of waste CFRP at large scales under microwave assisted pyrolysis, three parts of the samples in the crucible were studied via SEM and EDS as shown in Figure 4 (marked 1, 2, and 3). The effectiveness of resin removal was assessed through SEM images and with carbon and oxygen content determined using EDS element

analysis. Each RCF sample was peeled off from the outside, and its core used for SEM/EDS measurement. EDS analysis results of the samples are presented in Table 2.

**Figure 3.** Optical photo, SEM imagine, and EDS analysis: (**A**) CFRP waste of Giant Bicycles (GBW), (**B**) CFRPs waste sample (TF 650 ◦C) after treatment with traditional furnace pyrolysis at 650 ◦C, and (**C**) CFRPs waste (MW 650 ◦C) after treatment with micro-wave pyrolysis at 650 ◦C.

**Figure 4.** Photograph and SEM micrographs of the recycled carbon fibers in different part of ceramic crucible after microwave pyrolysis with varying condition; (**A**) microwave pyrolysis at 350 ◦C, (**B**) at 450 ◦C, (**C**) at 550 ◦C and (**D**) at 650 ◦C.


**Table 2.** EDS analysis of the samples.

a Each value is an average taken over 6 different spots on the samples.

At 350 ◦C pyrolysis temperature (Figure 4(A-1–A-3)), the A-1 area of SEM image shows the resin residue between the carbon fibers. EDS analysis determined the carbon,oxygen, and silicon content as 93.2 ± 0.7, 6.3 ± 0.6, and 0.5 ± 0.1%, respectively. Compared with samples GBW of Figure 3A, oxygen content decreased from 13.6 ± 4.7 to 6.3 ± 0.6% and carbon content increased from 85.8 ± 5.3 to 93.2 ± 0.7%. These results revealed that the resin is not fully pyrolyzed. In addition, the EDS and SEM results of A-2 and A-3 areas are similar. These SEM images show a significant quantity of resin adhered to the carbon fiber. Comparing the EDS results of three areas (Figure 4(A-1–A-3), Table 1), areas 2 and 3 show better resin removal. A similar situation also occurred under the different sample pyrolysis conditions applied.

At 450 ◦C (Figure 4(B-1–B-3), Table 1), the EDS results showed the carbon content is 99 ± 0.1, 99.6 ± 0.3, and 99.2 ± 0.2% and silicon content is 0.9 ± 0.1, 0.4 ± 0.3, and 0.8 ± 0.2%, respectively. Figure 4C,(D-1–D-3), shows the EDS and SEM results after microwave pyrolysis at 550 ◦C (B) and 650 ◦C (D). The carbon content of MW 550 ◦C from areas 1 to 3 is 98.9 ± 0.3, 99.1 ± 0.2, and 98.9 ± 0.3% and silicon content is 1.1 ± 0.3, 0.9 ± 0.2, and 1.3 0.3%, respectively. For MW650 ◦C, the carbon content is 99.2 ± 0.2, 99.4 ± 0.3, and 99.8 ± 0%, and the silicon content is 0.80.2, 0.6 ± 0.3 and 0.2 ± 0%, respectively. All of the above EDS results show that pyrolysis of waste CFRP is uniform at different crucible positions. While the pyrolysis condition is above 450 ◦C, the oxygen content reduced to 0%, indicating that the resin was fully carbonized. The difference between carbon content for the three areas is approximately 0.47 ± 0.01% at 550 and 650 ◦C. This was, perhaps, caused by the waste having a curved and irregular shape or non-uniform density. Verification of these statements requires further research. However, this pyrolysis condition had enough process time and temperature to reduce the influence caused by different crucible position.

The SEM images show filamentous matter on the fibers. The shape changed from the filament to a dot along with decreased silicon content. Section 3.4 will discuss silicon removal from CFRP waste through microwave pyrolysis.

#### *3.4. Silane Removal from CFRPs Waste at Different Microwave Pyrolysis Temperature*

Figure 5 shows a SEM diagram of CFRP materials that are thermally treated at different temperatures. A carbon fiber sample (pyrolysis 650) treated at 650 ◦C in a high-temperature oven shows significant resin filling, as shown in Figure 5A. The results of the EDS element analysis of the sample showed that the fiber surface of TF at 650 ◦C had an oxygen content of approximately 8.5 ± 1.2%. Based on results of the TGA analysis, the resin had been carbonized to form carbide; this means, for a traditional furnace, resin pyrolysis is not complete. For the residual resin layer wrapped on the surface of waste CFRPs, the Si element content is 0.3 ± 0.04%, while the CFRPs samples treated with microwave pyrolysis at 350 ◦C (Figure 5B), 450 ◦C (Figure 5C), and 550 ◦C (Figure 5D), show a cotton-like or filamentous substance on the surface of the carbon fibers. The silicon content is approximately 0.9 to 0.4%. At a treatment temperature of 650 ◦C (Figure 5E), the shape of filamentous substances changed to dots (silicon content: 0.2 ± 0%).

**Figure 5.** SEM micrographs of the recycled carbon fibers by thermally treated at different temperatures; (**A**) traditional furnace pyrolysis treated at 650 ◦C, (**B**) microwave pyrolysis treated at 350 ◦C, (**C**) at 450 ◦C, (**D**) at 550 ◦C, (**E**) at 650 ◦C; and (**F**) 950 ◦C.

By way of understanding the effect of microwave thermal pyrolysis temperature on residual silane content on waste CFRPs surface, we also set the temperature of microwave pyrolysis at 950 ◦C. Following this treatment, the silicon content is 0.1%. The SEM image shows that the recovered carbon fiber surface is completely free of residues or obvious defects, as shown in Figure 5F. In Figure 6, the MW 950 ◦C Figure 6A,B compared with the original carbon fiber (Mitsubishi TR 50, Figure 6C,D), demonstrates they are without defects on the fiber surface, which can be further proved by high magnification FE-SEM.

However, even if the microwave pyrolysis temperature increases by 300 ◦C, the silicon content is only slightly reduced. Therefore, the results show the silicon content is gradually reduced from the recovered carbon fiber along with breakdown of the resins, but it cannot be completely removed by microwave pyrolysis at 350 to 950 ◦C.

Since the organic components in the silane compounds typically degrade when heated to approximately 400 ◦C, only elemental silicon remains. Silicon gasification temperature is 3265 ◦C, which is difficult to achieve via thermal decomposition methods. Other researchers also presented SEM/EDS results of pyrolyzed carbon fiber pretreated using silane coupling

agen<sup>t</sup> that exhibit the same dots on the carbon fiber surface and were determined as containing elemental silicon [63,64].

**Figure 6.** High magnification FE-SEM micrographs of the recycled carbon fibers (MW 950 ◦C, (**A**); at 5000× magnification, (**B**); 10,000×) and the original carbon fiber (Mitsubishi TR 50, (**C**,**D**) they magnification ware respectively 5000× and 10,000×).

For these reasons, these filamentous substances are inferred to be residual silicon and carbonized resin produced during the CFRPs pyrolysis process. The residual silicon is removed as the carbonized resin pyrolyzed to form carbon dioxide or carbon monoxide by air atmosphere. Despite the evidence for the above conclusions, further research is required to affirm this interpretation.
