Recycling of composite waste by sequential application of multi-criteria decision-making,pyrolysis,and reproduction

Abstract

This paper summarizes our work aimed at applying different methods sequentially in the evaluation of GFRP (glass fiber reinforced polyester) waste, which is a composite waste according to circular economy assumptions. For this aim, the management alternatives that are defined as mechanical recycling, chemical recycling, pyrolysis, and incineration methods are compared using the ANP (Analytic Network Process) and the VIKOR (Vise Kriterijumska Optimizacija I Kompromisno Resenje) methods. To show the applicability of the pyrolysis, which was determined as the most appropriate method, experiments were carried out at different temperatures. After the general characterization of the products was conducted, the solid product was used to reproduce GFRP. After a tensile strength test of the GFRP, which was produced using fibers obtained from pyrolysis at 500°C, it was found that it had approximately 43.15% more tensile strength than polyester cast without fiber. Following a further oxidation step, the tensile strength increased by 61.50%. The liquid products have the potential to be used as fuel oil, and the gas product is sufficient to provide the energy requirements of a process plant. Pyrolysis proved to be a profoundly feasible method for not only recovering glass fiber, but also energy from GFRP waste.

Keywords

Analytic network process (ANP)glass fiber reinforced polyester (GFRP)multi-criteria decision-making (MCDM)pyrolysis vise kriterijumska optimizacija I kompromisno Resenje (VIKOR)

Introduction

Composites are materials that are obtained by combining the positive properties of multiple materials in one material or by combining them at a macro level to create a new feature. GFRP, which was developed in the late 1940s, is the most widely-used and the first modern polymer-based composite material. GFRP is a high-quality composite engineering material that is obtained by combining plastic with insufficient mechanical strength and glass fiber with high mechanical strength. As a result, the physical properties of plastics such as tensile strength, flexural strength, rigidity, and impact strength can be increased. The most common thermoset resins used in the composite industry are unsaturated polyesters, epoxies, vinyl esters, polyurethanes, and phenolics. Unsaturated polyester resin is the most important polymer in the composite industry, and it represents approximately 75% of the resins used.¹

The generation of serious waste for GFRP, the mass production of which started in the 1970s and continue to be used increasingly, started from the beginning of the 2000s. Over one million tons of thermoset composites are produced each year and GFRP is the main production among these composites.² The first problem with GFRP waste recycling is the use of thermoset resins. Thermoset resins are liquids that are solidified by a chemical reaction cross-linked to resin molecules. This reaction is irreversible; the cooked solid resin does not return to its original liquid by heating, as with thermoplastics. Unlike thermoplastics, GFRP waste cannot be melted or kneaded.³

There are successful applications for thermoset composites in the literature, but it has been pointed out that recycling of the product is a challenging problem after it is converted to waste. In addition, it is stated that the problems in recycling increase their importance and this problem prevents the development of thermoset products, even causing them to not be used in certain sectors.4 As can be seen from Figure 1, mechanical, chemical, and thermal methods can be used for GFRP waste management.

Figure 1.

GFRP waste management methods (adopted from Kostopoulos et al., 2008.⁵

There have been studies on the use of waste as filling and fiber material as mechanical methods. Bledzki et al. (1999),⁶ Bartholomew (2004),³ Palmer (2009),⁷ and Yazdanbakhsh et al. (2018)⁸ have worked on the use of waste as a filling material in the building sector. Derosa and Telfeyan (2005),⁹ Kostopoulos et al. (2008),⁵ Rahimizadeh et al. (2021)¹⁰ and Pickering (2006)¹¹ investigate the potential of using fibers obtained from waste instead of raw fiber in different applications. In fibers obtained by mechanical recycling; undesirable properties such as low density, single fiber, variable fiber length, and decreased fiber strength are observed.¹²

The purpose of chemical methods is to separate the fibers from the resin and to make them reusable with the help of a chemical. For this purpose, maleic anhydrides,¹³ acids, alkalis,^5,14,15 and organic solvents such as dichloromethane, ethanol, acetone,^16,17 and subcritical water¹⁸ are used. The common idea in these studies is that the quality of the fiber obtained from the chemical conversion is reduced. Additionally, chemical recycling has limited commercial applications, with large investment costs and potential for significant waste streams.⁹

The two basic methods of thermal methods are pyrolysis and incineration. Kostopoulos et al. (2008)⁵ argues that pyrolysis is the best method in thermal processes with regard to GFRP recycling, while Rahimizadeh et al. (2020)¹⁹ states that while pyrolysis can significantly degrade recovered fibers. According to Wong et al. (2017),²⁰ compared to other commonly adopted industrial processes, energy consumption (30 MJ/kg) of the pyrolysis process is moderate. 400–700°C is generally used for the pyrolysis temperature. It has been stated that the presence of aromatics is high in the liquid products, CO and CO₂ are present in the gas product, glass fibers are recovered in the solid product and that these glass fibers can be purified by oxidation.^11,21,22

Pickering (2006)¹¹ states that incineration with energy recovery is also a viable option for GFRP waste due to its high calorific value. He also states that if coal and waste composites are burned together in a fluidized bed incinerator, the filler calcium carbonate in the composites will absorb sulfur oxides from the combustion of the coal, thereby reducing sulfur emissions. However, air emissions, metal, and fiber contents in ash are also seen as disadvantages of incineration.³

As can be seen, there are different methods for the management of GFRP waste and they have both advantages and disadvantages. In studies on the general evaluation of these methods, different factors that should be considered have been identified.^23–28 From this point on, MCDM (Multi-Criteria Decision-Making) is a good option in evaluating all the methods by considering various factors.

MCDM techniques are useful tools for applications designed to solve multiple-choice problems. In the literature, there are several studies related to waste management options that are given in Table 1.

Table 1.

MCDM applications on waste management in the literature.

Application field	MCDM method	Reference
Bioenergy production technology alternatives	COPRAS^a	Ref. 29
End-of-life alternatives for waste plastics	MAVT^b	Ref. 30
Waste-to-energy technologies	TOPSIS^c	Ref. 31
Treatment of oil-contaminated soils	BWM-MCDM^d	Ref. 32
Leachate treatment options	ANP^e and PROMETHEE^f	Ref. 33
Health-care waste treatment technology	TOPSIS^c	Ref. 34
Options for energy recovery from municipal solid waste	AHP^g	Ref. 35
The location for hazardous waste treatment facilities	SAW^h	Ref. 36
The municipal landfill site selection	ANP^e	Ref. 37

^aComplex proportional assessment.

^bMulti-attribute value theory.

^cTechnique for order preference by similarity to ideal solution.

^dBest-worst multi-criteria decision-making.

^eAnalytic network process.

^fPreference ranking organization method for enrichment evaluations.

^gAnalytical hierarchy process.

^hSimple additive weighting.

On the other hand, there is just one study related to MCDM for waste composite materials; mechanical recycling, pyrolysis, microwave pyrolysis, and solvolysis for carbon and glass fiber reinforced plastic wastes were evaluated by the AHP and the TOPSIS methods.³⁸ Contrary to this, in this study, the ANP and the VIKOR methods are used with different criteria and methods. At the end of the MCDM studies, the most appropriate method was found to be pyrolysis. To show the applicability of the pyrolysis, pyrolysis experiments were conducted with GFRP waste at various temperatures. Finally, a composite material with the solid product obtained from pyrolysis was produced and its properties were determined.

Materials and methods

Figure 2 shows the flowchart of the study. This study includes three basic stages: ANP and VIKOR studies, testing the applicability of the selected method, and reproduction of the composite material.

Figure 2.

Flowchart of the study.

Problem definition

In this study, the aim was to recover glass fiber from GFRP waste, which is difficult to recycle. As an example, a 1.5 mm thick GFRP waste plate, which is a transparent panel used for roofing and siding with a glass fiber content of 25 wt %, was used (Figure 3). It is estimated that the waste occurs after approximately 30 years of use. Its weight is about 2 kg/m² and its density is 1.2 g/cm³. According to the information obtained from the manufacturer, the tensile strength of the unused GFRP was 52 MPa.

Figure 3.

GFRP waste sample.

The ANP and the VIKOR application

In this study, the ANP and VIKOR were preferred as the MCDM method because of the different approach involved. In ANP studies, binary comparisons are conducted to assess the relative importance of criteria and alternatives. Pairwise comparison matrices are constructed in the ANP using the 1–9 relative importance scale proposed by Saaty.³⁹ To verify the consistency of the judgments of the pairwise comparison matrices, the consistency ratio of the matrices is calculated in a similar way to the AHP method. The consistency ratio should be less than or equal to 0.1.⁴⁰ At the last stage, the importance weights of alternatives, factors, and sub-factors can be determined from the limited super matrix. The highest importance weight indicates the best alternative.⁴¹

In this study, a BCR (Benefit-Cost-Risk) model was created. First, criteria and alternatives are defined. After this, certain criteria are presented as strategic criteria for evaluating the priorities of the BCR model. The pairwise comparisons during the assessment process are based on literature surveys, the stakeholders' experience, and common knowledge. The weights of the benefits (b), costs(c), and risks(r) models are determined as 1/2, 1/6, and 1/3, respectively. The priorities of the alternatives are obtained by combining the results from the benefits, costs, and risks models using a reciprocal formula ($NormalNet (Benefits) * $SmartAlt (Benefits) + $NormalNet (Costs) * $SmartInvAlt (Costs)+$NormalNet(Risks)*$SmartInvAlt(Risks)). In this formula, Normal means normalized so that the values in the vector sum to 1.000 and Smart means which uses the best of the three (ideal, normal, total) for a given situation.

The VIKOR is a method developed by Opricovic (1998) for multi-criteria optimization of complex systems. Closeness to the ideal solution is a key part of the VIKOR method. Ideal alternatives are comparisons according to a closeness measure.⁴² The steps of the VIKOR method are listed as follows:

• Finding the Best (fj+) and Worst (fj−) Values

• Creation of the normalization matrix

• Calculation of the weighted normalization decision matrix

• Calculation of Si and Ri Values

• Calculation of the VIKOR index (Qi)

• Determining the final ranking of alternatives

• Determination of acceptable advantage and acceptable stability clusters

Four alternatives were evaluated using the ANP and VIKOR methods in this study: mechanical recycling, chemical recycling, pyrolysis, and incineration. According to a literature survey, alternatives and explanations are given in Table 2.

Table 2.

Alternatives used in the study.

Alternatives	Explanations
Mechanical recycling	This method is basically cutting waste GFRP materials into smaller pieces, grinding them, and then sizing the different pieces. Mechanical recycling products are considered as powder filling material if the grain size is less than 0.5 mm, and glass fiber if it is between 0.5 mm and 2 mm. The obtained products may be used in the construction sector such as insulation material and filling material, as well as being molded with new resins to obtain GFRP again. However, the new FRPs obtained generally have lower strength.⁷
Chemical recycling	In the chemical recycling method, fibers and plastic materials can be recovered by dissolving the polyester in the waste GFRP with solvents. In a study in which subcritical water was used in the solvolysis process applied to dissolve polyester resin and recover glass fiber, the effect of recovered glass fiber on mechanical performance was examined and it was determined that the fibers did not lose their mechanical properties. In addition, it was determined that DCM (dichloromethane) is an effective solvent for chemical recycling with GFRP. However, it was determined that mechanical degradation increases with increasing temperature.¹⁵
Pyrolysis	Pyrolysis is a thermochemical process in which organic materials undergo thermal cracking at high temperatures and in an oxygen-free environment. At the end of the pyrolysis process, organic materials decompose as liquid and gas products. It is possible to obtain inorganic fibers with the pyrolysis process and energy can be obtained from the liquid and gas products with high calorific values. It has been found that fiber can be recovered by pyrolysis, but thermal pressures cause surface defects that reduce the strength of the recovered fibers.⁵ The gross calorific value of the gas product varies depending on the temperature and it has been stated that it can have a calorific value of more than 30 MJ/Nm³ at 500, 550, and 600°C.²¹ The liquid product obtained can be used as fuel like petroleum.⁴³
Incineration	Waste GFRPs are generally incinerated after reducing their size. Incineration can be performed directly by burning GFRPs or by using 10% GFRP in waste incineration systems. The calorific value of the resins has been determined to be approximately 30000 kJ/kg.¹¹ Micronized carbon particles, PAHs (polycyclic aromatic hydrocarbons), carbon monoxide, and carbon dioxide that causes global warming are released during incineration. Fibers and fillers remaining after burning can be used as raw materials in cement plants.⁴⁴

In this study, a total of 13 criteria are considered, and the properties of these criteria are summarized in Table 3.

Table 3.

Criteria.

	Name of criteria	Remarks
Benefit	Glass fiber production	Glass fiber recycling potential was evaluated in the alternatives.
	Recycling of plastic material	The recycling potential of plastics in the composite material was evaluated.
	Energy recovery	Energy recovery potential was evaluated in the alternatives.
	Process time	Alternatives with shorter processing times were evaluated positively.
Cost	Equipment cost	In this criterion, the costs of tools and equipment used in the process were considered.
	Plant cost	In this criterion, the costs of the plant used in the process excluding equipment were considered.
	General operating cost	In this criterion, the costs of energy requirement per kg, personnel, and transportation used in the process were considered.
	Analysis cost	In this criterion, the costs of periodical and continuous measurements were taken in terms of process safety and environmental legislation in determining the quality of the products and the fuels obtained from the process.
Risk	Residual waste	Composite parts, waste solvent, ash, and suchlike materials with no economic value, except for fuel and filler material obtained were evaluated within this scope.
	VOC and toxic gas release	During the processes, VOCs (volatile organic compounds) and toxic gas components, due in particular to resins and solvents were evaluated.
	The loss of tensile strength of glass fiber	The strength that the recovered fiber had lost compared to virgin fiber.
	Industrial accident	Industrial accidents potential such as flash, explosion, and fire caused by used chemicals, supplemental fuel, obtained gas, and liquid products were considered.
	Commercialization potential	The chemical method was considered to have lower commercialization potential than others.

A scoring method for the criteria through pairwise comparison was applied, and the weights of criteria (%) are given in Table 4. For the VIKOR method, the values for the ANP have been normalized by considering the weights of benefit, cost, and risk clusters.

Table 4.

Weights of criteria.

Cluster	Criteria	Weights for the ANP (%)	Weights for the VIKOR (%)
Benefit	Glass fiber production	44.03	22.02
	Recycling of plastic material	23.31	13.99
	Energy recovery	27.97	11.66
	Process time	4.69	2.35
Cost	Equipment cost	11.38	1.82
	Plant cost	22.70	3.63
	General operating cost	61.24	9.80
	Analysis cost	4.68	0.75
Risk	Residual waste	6.42	2.18
	VOC and toxic gas release	30.48	10.36
	The loss of tensile strength of glass fiber	15.22	5.17
	Industrial accident	30.48	10.36
	Commercialization potential	17.40	5.92

GFRP characterization

As can be seen in the results section, since the most suitable method according to the ANP and VIKOR results is found to be pyrolysis, the GFRP waste plate was sheared to a size of 20–30 mm for evaluation by pyrolysis. The proximate analyses (moisture, ash, volatile matter, and fixed carbon content) of the analytical sample were determined. The elemental composition (C, H, N, and S) of the sample was determined using a Leco TruSpec CHN and S (USA).

Testing the applicability of the selected method

When studies in the literature are examined, it can be seen that pyrolysis temperatures vary between 300 and 1000°C, but that the most effective degradation is between 400 and 600°C.²⁷ The pyrolysis experiments were carried out under atmospheric pressure at temperatures of 400, 500, and 600°C with a 10°C/min heating rate in a fixed bed stainless steel (380 S) 240 cm³ reactor that was well-swept with nitrogen before pyrolysis. All of the experiments were carried out in typical runs where 10 g of a sample was placed in the reactor and kept there for 1 h after the reactor had reached the target temperature. The liquid phase was collected in a cold trap kept at approximately 0°C using ice. The solid and liquid product yields were determined gravimetrically, while the gas yields were determined by mass difference. All of the yields were expressed on a dry ash-free basis using the average yields of at least three experiments.

The elemental composition of the solid and liquid products was determined using a LECO TruSpec CHN and S, and the HHVs (Higher Heating Values) of liquid products were calculated using the weight percentage of the elements according to the Dulong equation Eq. (1).⁴⁵

Heating value (\frac{B t u}{1 b}) = (145 * C) + [610 * (H - (\frac{O}{8}))] + (40 * S) + (10 * N)

(1)

For post-pyrolysis in the presence of air, 450–600°C has been defined as the appropriate temperature range for the oxidation of pyrolytic char and for cleaning the fiber surface.^20,27 In this study, the organic residue in the solid products was removed by burning in a muffle furnace at 500°C for 30 min.

Reproduction of composite material

Glass fiber obtained by the pyrolysis of GFRP at different temperatures was remolded with ortho-phthalic polyester resin by the hand lay-up method. A gel coat was applied by brush on the surface of a rectangular mold of 2 mm depth. After the gel coat hardened, the reinforcing glass fiber 25% (by weight) was molded. Polyester resin mixed with 2% methyl ethyl ketone (MEK) was poured onto the fiber, dispersed by brush, and air bubbles were removed. This process was repeated after each coat application of the reinforcing material. The mold was covered and left for 4 h to harden at room temperature. To compare tensile strengths, the polyester resin was also molded without fiber reinforcement.

The tension bars were cut from the produced materials according to TS EN ISO 527-4 (Test conditions for isotropic and orthotropic fiber-reinforced plastic composites) as shown in Figure 4 and the dimensions given in Table 5.

Figure 4.

The shape of the tension bar.

Table 5.

Tension bar dimensions.

		Dimensions (mm)
L1	Total length	≥150
L2	Jaw distance	90 ± 0.5
L3	Length of the narrow section	60 ± 0.5
b1	Width of the narrow section	10 ± 0.2
b2	Width at the ends	20 ± 0.2
h	Thickness	2–10
R	Radius	≥60

First, three samples were prepared with the non-oxidized solid product obtained from pyrolysis studies carried out at 400, 500, and 600°C temperature, considering minimum energy consumption, then a new sample was produced by oxidizing the pyrolysis solid product with the best result (Figure 5). A glass fiber-free sample was prepared as a reference sample.

Figure 5.

Samples of reproduced composite materials.

Each produced material was fixed to the grips of the tensile testing device and tested in tension at 2 mm/min in an INSTRON 5581 Mechanical Test Device until failure. This test was carried out three times for each sample. A stress-strain curve was obtained during the tension test using BlueHill Software.

Results and Discussion

The results are given following sections according to the flowchart in Figure 2.

The ANP and VIKOR results

The normalized ANP results are given in Table 6. Pyrolysis was the best method in the benefit cluster because of its potential for both energy and material recovery. Although mechanical recycling is the primary method in the cost and risk clusters, it was ranked third overall, close to chemical recycling. It is noteworthy that incineration is placed last in all of the clusters. According to the overall results, it can be seen that the most appropriate method is pyrolysis, followed by chemical recycling, mechanical recycling, and incineration, respectively.

Table 6.

The ANP results.

Alternatives	Benefit		Cost		Risk		Overall
	Ratio	Rank	Ratio	Rank	Ratio	Rank	Ratio	Rank
Mechanical recycling	0.2198	3	0.1675	1	0.1792	1	0.2402	3
Chemical recycling	0.2806	2	0.2769	3	0.3474	4	0.2612	2
Pyrolysis	0.3578	1	0.2366	2	0.1897	2	0.3343	1
Incineration	0.1418	4	0.3190	4	0.2838	3	0.1644	4

The S_i, R_i, Q_i values and the rankings of the alternatives are given in Table 7. According to the results of the VIKOR method, pyrolysis is the best ranked by Q_i, as in the ANP. Mechanical recycling, chemical recycling, and incineration come after pyrolysis, respectively.

Table 7.

S_i, R_i, Q_i values and rankings of the alternatives.

Alternatives	S_i	R_i	Q_i	Rank
Mechanical recycling	22.22	11.66	0.2205	2
Chemical recycling	54.53	22.02	0.5054	3
Pyrolysis	21.60	3.89	0.0000	1
Incineration	71.96	10.36	1.0000	4

The case related to the acceptable advantage and acceptable stability of the obtained results was checked and it was determined that both conditions were met. For condition 1, the difference of A₁ (best alternative) and A₂ (second best alternative) is less than 1/(m-1) (Here, “m” is the number of alternatives). For condition 2, the A₁ (pyrolysis) alternative with the best Q value has achieved the best score of both the S and R values.

GFRP characterization results

The results of proximate analysis and elemental analysis of the GFRP are shown in Table 8. Pyrolysis is a feasible process since the volatile matter is the main component of GFRP because of plastic content. Compared to other plastics, GFRP has a high ash content (26.3%) because the polymer matrix is reinforced with fiber. Since the hydrogen content of GFRP (64.28%) is relatively high, hydrogen will evolve during pyrolysis.⁴⁶

Table 8.

The proximate and elemental analysis results of the GFRP.

Content		ASTM No	%
Moisture		D-3173-85	0.95
Ash		D-3174-82	26.53
Volatile matter		D-3172-73	72.52
Fixed carbon		D-3175-82	<0.001
Elemental	C	—	49.32
	H	—	4.28
	O^a	—	18.92
	N	—	<0.001
	S	—	<0.001

^aIt was calculated from the difference considering the moisture and ash values.

Pyrolysis results

The yields of the products obtained by the pyrolysis of the GFRP at 400, 500, and 600°C are given in Table 9. While the solid product yield decreased with increasing temperature, the liquid and gas product yields increased. The highest liquid product yield was obtained at 500°C. It was seen that the degradation of the polymer matrix was complete up to 500°C, as there was no further significant increase in total yield of liquid and gas. At higher temperatures, gas efficiency increases due to the increased cracking of volatiles, with a resulting decrease in oil yield.⁴⁷ According to the published literature, in GFRP pyrolysis, the decomposition of the resin is completed between 450 and 500°C.^19,43 At 600°C, the liquid product yield decreased while the gas product yield increased. Unlike other waste materials used in pyrolysis (plastic, biomass, and others), due to the wide structural changes during the production of composite materials, a wide range of product yields have been observed in the literature.

Table 9.

Pyrolysis product yields of GFRP (dry ash-free).

Reference	Temperature (°C)	Solid product (%)	Liquid product (%)	Gas product (%)
This study	400	21.32	34.47	44.21
	500	19.79	55.69	24.52
	600	19.18	30.01	50.81
18	—	38.3–84.7	5.1–64.1	0.2–14.4
39	—	72–82	9–12	6–12
19	—	≈20	≈40	≈40

Before the obtained solid products were oxidized, elemental analyzes were carried out (Table 10). It can be said that the degradation is still incomplete due to the high carbon content in the solid product at 400°C. It is also seen that the lowest carbon content stuck to the fiber is at 500°C. Figure 6 shows the glass fiber recovered from subsequent oxidation of the solid product of pyrolysis at this temperature. The separated fiber did not tend to agglomerate or stick together. Most of the fibers were not broken.

Table 10.

Elemental analysis result of the pyrolysis solid products.

Elemental (%)	400°C	500°C	600°C
C	21.93	6.41	8.76
H	0.19	0.13	<0.001
N	<0.001	<0.001	<0.001
S	<0.001	<0.001	<0.001

Figure 6.

Recovered glass fiber after oxidizing.

Table 11 shows the elemental analysis result and the Higher Heating Values (HHV) of pyrolysis liquid products obtained at 400, 500, and 600°C. Pyrolysis oils obtained at lower temperatures contain less carbon and more oxygen than the oil obtained at 600°C. This can be attributed to the fact that the more oxygen-containing functional group remains attached to the pyrolysis fluids molecules, as less cracking occurs at low temperatures.⁴⁸ The heating value of the liquid product was low in comparison to the conventional hydrocarbon liquid fuels, due to its high oxygen content, but sufficiently high to make it a viable fuel.²⁰ Sokoli et al. (2016)⁴⁹ obtain a liquid with a higher heating value of 37.3 MJ/kg as a result of GFR unsaturated polyester resin pyrolysis, and they find that the liquid oil was similar to crude oil in terms of its HHV (44.6 MJ/kg) and elemental composition, and therefore, have the potential to be used as fuel oil.

Table 11.

Elemental analysis and HHV results of the pyrolysis liquid products.

		400°C	500°C	600°C
Elemental (%)	C	67.89	66.66	70.36
	H	6.11	6.49	6.74
	O ^a	26.00	26.85	22.9
	N	<0.001	<0.001	<0.001
	S	<0.001	<0.001	<0.001
HHV (MJ/kg)		26.9	26.9	29.2

^acalculated from the difference.

In terms of gas products, the data in the literature show that the gas also has a content which can be evaluated. Cunliffe et al. (2003)⁵⁰ obtain a gas product with a 14.5 MJ/m³ calorific value consisting of 62.3% carbon dioxide, 2.8% carbon monoxide, 9.3% hydrogen, 10.5% ethane, and ethylene by pyrolyzing polyester resin composite waste at 500°C. Torres et al. (2000)⁴³ pyrolyze a fiberglass reinforced ortho-phthalic polyester at 500°C and obtain a gas product with a 14.6 MJ/m³ calorific value consisting of 58.9% carbon dioxide, 33.4% carbon monoxide, and 7.7% C₁-C₄.

Tensile strength test results

The stress-strain curves are given in Figure 7 and the tensile strengths of the samples are given in Table 12. According to the figure, it is seen that the extensions are similar in GFRPs reproduced with the solid product at 500°C and oxidized solid product, and similar in reproduced GFRPs with the solid product at 400 and 600°C. It has been found that the recovered glass fiber with the pyrolysis of GFRP at 500°C increases the strength of polyester the most. This is thought to be due to the low carbon content of the solid product at 500°C. After this, the tensile strength was found to be 17.28 MPa in the composite material obtained by oxidizing the product at 500°C. Approximately, a13% improvement in tensile strength was achieved with the oxidation process. Similarly, it was found that the tensile strength for reproduced GFRP with non-additive, non-oxidized, and oxidized fibers after pyrolysis was 11 MPa, 15.3 MPa, and 18.5 MPa by Onwudili et al. (2016).⁵¹ Kouparitsas et al. (2002)⁵² recovered short fibers from polyester-based glass fiber composites by mechanical recycling and prepared a new thermoplastic composite with polypropylene at a content level of 40 wt %. They report that the tensile strength of the new composite is 25 MPa. Pickering et al. (2000)⁴⁴ also find that strength is not only dependent on fiber strength, but also on fiber length and binder content.

Figure 7.

Stress-strain curves of the samples.

Table 12.

Tensile strength values of the samples.

Sample	Tensile strength (Mpa)
Polyester without fiber	10.70 ± 0.99
Reproduced GFRP with the solid product at 400°C	13.15 ± 0.22
Reproduced GFRP with the solid product at 500°C	15.32 ± 0.43
Reproduced GFRP with the solid product at 600°C	12.88 ± 0.02
Reproduced GFRP with the oxidized solid product	17.28 ± 0.17

Considering that the tensile strength of unused GFRP is 52 MPa, the strength of the composite produced with recovered fibers is low. However, light-duty parts, such as vehicle headlight housings and instrument panels, offer significant potential for commercial applications. Torres et al. (2009)⁵³ also report that pyrolysis solids can be recycled in a proportion of 6 wt % in a bulk molding compound to replace virgin filler and fiberglass, with no damaging effect on mechanical properties. Cunlife et al. (2003)⁴⁷ find that glass fiber recovered from solid residue could replace up to 20 wt % of virgin glass fiber in a dough molding compound.

Conclusion

For the recycling of GFRP waste used as roofing material; mechanical recycling, chemical recycling, pyrolysis, and incineration methods were examined using the Multi-Criteria Decision-Making technique. Of these four processes, pyrolysis was found to be the most convenient method in both of the ANP and the VIKOR methods. To show the applicability of the pyrolysis, pyrolysis experiments were conducted with GFRP waste at different temperatures, with solid, liquid, and gas products being obtained. The major results and conclusions regarding the observed criteria are summarized as follows:

(1) The solid residue mostly consisted of glass fiber and a certain amount of carbonaceous material.

(2) Recycling of plastic material was achieved through the transformation of plastic material into liquid and gaseous products.

(3) Glass fiber production from the solid residue was conducted and has been successfully used in reproduction. The strength test of the reproduced GFRPs show that the solid product obtained after pyrolysis is usable. After the tensile strength test of the GFRP, which was produced by using fibers obtained from pyrolysis at 500°C, it was found that it had approximately 43.15% more tensile strength than polyester cast without fiber. With a further oxidation step, tensile strength increased by 61.50%. Glass fiber recovered from solid residue can replace virgin fiber in certain proportions in composite molding. In cases where aesthetics is not important and strength is sufficient, fiberglass can also be used in the production of composites without further oxidation.

(4) It was found that the liquid product, created by the pyrolysis process at different temperatures differed quantitatively. However, when the elemental analysis was conducted, they had similar content. It was found that the carbon content varied by approximately 4% and the hydrogen content varied by approximately 0.5%. The heating values of the liquid product obtained after pyrolysis at 400°C, 500°C, and 600°C were calculated as 26.9 MJ/kg, 26.9 MJ/kg, and 29.2 MJ/kg, respectively. Therefore, liquid products have the potential to be used as a fuel oil for energy recovery.

(5) The data in the literature show that the gas product with 14.5 MJ/m3 calorific value mostly consisting of carbon dioxide, carbon monoxide, and C1-C4, is sufficient to provide the energy requirements of a process plant.

Pyrolysis proves to be a profoundly feasible method, not only for the recovery of glass fiber, but also for energy from GFRP waste. In future studies, life cycle assessment and life cycle cost of GFRP recycling by pyrolysis may be evaluated from an environmental and economic perspective.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Aysun Özkan

References

Composites Institute . SPI/CI introduction to composites. USA: CRC Press, 1998.

Quadrini

Bellisario

Santo

L. J

. Molding articles made of 100% recycled fiberglass. Compos Mater 2016; 50(21), 2959–2969. DOI: 10.1177%2F0021998315615199

Bartholomew

. Fiberglass reinforced plastics recycling. Dec. Report, MnTAP, 2004.

Sims

Bishop

. UK polymer composites sector- Foresight Study and Competitive Analysis. Ireland: Manufacturing of Advanced Composites, 2001.

Kostopoulos

Fotiou

Loutas

. Recycling and reuse of glass-polyester composite materials. An experimental study. In: Proceedings of the polymer processing society 24th annual meeting-PPS-24, Salerno, Italy, 15 June - 19 June 2008.

Bledzki

Sperber

Theis

, et al. Holzgefüllte thermoplaste als alternative fir naturholz. Gummi, Fasern, Kunststoffe 1999; 52(4), 294–296.

Palmer

Mechanical recycling of automotive composites for use as reinforcement in thermoset composites. PhD dissertation, UK: University of Exeter, 2009.

Yazdanbakhsh

Bank

Tian

. Mechanical processing of GFRP waste into large-sized pieces for use in concrete. Recycling 2018; 3(1), 2–11. DOI: 10.3390/recycling3010008

Derosa

Telfeyan

. Current state of recycling sheet molding compounds and related materials. J Thermoplas Comp Mat 2005; 18(3), 219–240. DOI: 10.1177%2F0892705705045006

10.

Rahimizadeh

Kalman

Fayazbakhsh

, et al. Mechanical and thermal study of 3D printing composite filaments from wind turbine waste. Polym Compos 2021; 42(5):2305–2311. DOI: 10.1002/pc.25978

11.

Pickering

. Recycling technologies for thermoset composite materials current status, Composites: Part A 2006; 37, 1206–1215. DOI: 10.1016/j.compositesa.2005.05.030

12.

Turner

Warrior

Pickering

. Development of high value moulding compounds from recycled carbon fibres. Plast Rubber Compos 2010; 39(3–5), 151–156. DOI: 10.1179/174328910X12647080902295

13.

Yoon

. Developing an extracting method of laminated glass-fiber for waste FRP boats regenerating. J Korean Soc Mar Environ Energy 2008; 11(1), 50–54.

14.

Dang

Kubouchi

Sembokuya

, et al. Chemical recycling of glass fiber reinforced epoxy resin cured with amine using nitric acid. Polymer 2005; 46: 1905–1912. DOI: 10.1016/j.polymer.2004.12.035

15.

Zabihi

Ahmadi

Liu

, et al. Sustainable approach to the low-cost recycling of waste glass fibres composites towards circular economy. Sustainability 2020; 12(2):641. DOI: 10.3390/su12020641

16.

Oliveux

Bailleul

La Salle , et al. Chemical recycling of glass fibre reinforced composites using subcritical water. Composites: Part A 2012; 43(11), 1809–1818. DOI: 10.1016/j.compositesa.2012.06.008

17.

Okajima

Sako

. Recycling of carbon fiber-reinforced plastic using supercritical and subcritical fluids. Mater Cycles Waste Manage 2017; 19, 15–20. DOI: 10.1007/s10163-015-0412-9

18.

Kao

Ghita

Hallam

, et al. Mechanical studies of single glass fibres recycled from hydrolysis process using sub-critical water. Composites: Part A 2012; 43, 398–406. DOI: 10.1016/j.compositesa.2011.11.011

19.

Rahimizadeh

Tahir

Fayazbakhsh

, et al. Tensile properties and interfacial shear strength of recycled fibers from wind turbine waste. Composites A 2020; 1(131), 105786. DOI: 10.1016/j.compositesa.2020.105786

20.

Wong

Rudd

Pickering

, et al. Composites recycling solutions for the aviation industry. Sci China Technol Sci 2017; 60(9), 1291–1300.

21.

Cunliffe

Williams

. Characterisation of products from the recycling of glass fibre reinforced polyester waste by pyrolysis. Fuel 2003; 82(18), 2223–2230. DOI: 10.1016/S0016-2361(03)00129-7

22.

Giorgini

Leonardi

Mazzocchetti

, et al. Pyrolysis of fiberglass/polyester composites: recovery and characterization of obtained products. FME Trans 2016; 44, 405–414.DOI: 10.5937/fmet1604405G

23.

Vijay

Rajkumara

Bhattacharjee

. Assessment of composite waste disposal in aerospace industries. Proced Environ Sci 2016; 35, 563–570. DOI: 10.1016/j.proenv.2016.07.041

24.

Nilakantan

Nutt

. Reuse and upcycling of aerospace prepreg scrap and waste. Reinforced Plastics 2015; 59(1), 44–51. DOI: 10.1016/j.repl.2014.12.070

25.

Gharde

Kandasubramanian

. Mechanothermal and chemical recycling methodologies for the Fibre Reinforced Plastic (FRP). Environ Technol Innovat 2019; 14, 100311–100341. DOI: 10.1016/j.eti.2019.01.005

26.

Gopalraj

Karki

. A review on the recycling of waste carbon fibre/glass fibre‐reinforced composites: fibre recovery, properties and life‐cycle analysis. SN Appl Sci 2020; 2:433–454. DOI: 10.1007/s42452-020-2195-4

27.

Naqvi

Prabhakara

Bramer

, et al. A critical review on recycling of end-of-life carbon fibre/glass fibre reinforced composites waste using pyrolysis towards a circular economy. Resour Conserv Recycl 2018; 136, 118–129. DOI: 10.1016/j.resconrec.2018.04.013

28.

Kalkanis

Psomopoulos

Kaminaris

, et al. Wind turbine blade composite materials- End of life treatment methods. Energy Proced 2019; 157, 1136–1143. DOI: 10.1016/j.egypro.2018.11.281

29.

Mishra

A. R.

Rani

Pandey

, et al. Novel multi-criteria intuitionistic fuzzy SWARA–COPRAS approach for sustainability evaluation of the bioenergy production process. Sustainability. 12(10), 4155 (2020) DOI: 10.3390/su12104155

30.

Deshpande

Skaar

Brattebø

, et al. Multi-criteria decision analysis (MCDA) method for assessing the sustainability of end-of-life alternatives for waste plastics: a case study of Norway. Sci Total Environ 2020; 719, 137353. DOI: 10.1016/j.scitotenv.2020.137353

31.

Alao

Ayodele

Ogunjuyigbe

ASO

, et al. Multi-criteria decision based waste to energy technology selection using entropy-weighted TOPSIS technique: the case study of Lagos, Nigeria. Energy 2020; 201, 117675. DOI: 10.1016/j.energy.2020.117675

32.

Kujlu

Moslemzadeh

Rahimi

, et al. Selecting the best stabilization/solidification method for the treatment of oil-contaminated soils using simple and applied best-worst multi-criteria decision-making method. Environ Pollut 2020; 263, 114447. DOI: 10.1016/j.envpol.2020.114447

33.

Özdemir

Özkan

Günkaya

, et al. Decision-making for the selection of different leachate treatment/management methods: the ANP and PROMETHEE approaches. Environ Sci Pollut Res 2020; 27(16), 19798–19809. DOI: 10.1007/s11356-020-08524-7

34.

You

Liu

, et al. Health-care waste treatment technology selection using the interval 2-tuple induced TOPSIS method. Int J Environ Res Public Health 2016; 13(6), 562. DOI: 10.3390/ijerph13060562

35.

Yap

Nixon

. A multi-criteria analysis of options for energy recovery from municipal solid waste in India and the UK. Waste Manage 2015; 46, 265–277. DOI: 10.1016/j.wasman.2015.08.002

36.

Sauri-Riancho

Cabanas-Vargas

Echeverría-Victoria

, et al. Locating hazardous waste treatment facilities and disposal sites in the State of Yucatan, Mexico. Environ Earth Sci 2011; 63(2), 351–362. DOI: 10.1007/s12665-010-0706-0

37.

Banar

Kose

Ozkan

, et al. Choosing a municipal landfill site by analytic network process. Environ Geology 2007; 52(4), 747–751. DOI: 10.1007/s00254-006-0512-x

38.

Delvere

Iltina

Shanbayev

. et al. Evaluation of polymer matrix composite waste recycling methods. Environ Clim Technol 2019; 23(1), 168–187. DOI: 10.2478/rtuect-2019-0012

39.

Saaty

. Decision making with the analytic hierarchy process. Int J Serv Sci 2008; 1(1):83–98. DOI: 10.1504/IJSSci.2008.01759

40.

Saaty

. Decision making with dependence and feedback: the analytic network process. 2nd ed. Pittsburgh, USA: RWS Publications, 4922 Ellsworth Avenue, 2001.

41.

Razavi Toosi

Samani

JMV

. Evaluating water transfer projects using analytic network process (ANP). Water Resour Manage 2012; 26, 1999–2014. DOI: 10.1007/s11269-012-9995-2

42.

Opricovic

. Multicriteria optimization of civil engineering systems. Belgrade: Faculty of Civil Engineering, 1998; vol. 2(1), pp. 5–21.

43.

Torres

de Marco

Caballero

, et al. GC-MS analysis of the liquid products obtained in the pyrolysis of fibre-glass polyester sheet moulding compound. J Anal Appl Pyrolysis 2001; 58, 189–203. DOI: 10.1016/S0165-2370(00)00122-4

44.

Pickering

Kelly

Kennerley

, et al. A fluidised-bed process for the recovery of glass fibres from scrap thermoset composites. Compos Sci Technol 2000; 60(4), 509–523. DOI: 10.1016/S0266-3538(99)00154-2

45.

Tchobanoglous

. Solid waste management. Environmental engineering: environmental health and safety for municipal infrastructure, land use and planning, and industry. New Jersey: Wiley, pp. 177–307, 2009.

46.

Yun

Seo

Koo

, et al. Pyrolysis characteristics of GFRP (glass fiber reinforced plastic) under non-isothermal conditions. Fuel 2014; 137, 321–327. DOI: 10.1016/j.fuel.2014.08.001

47.

Williams

. Recycling plastic waste by pyrolysis. J Inst Energy 1998; 71, 81–93.

48.

De Marco

Legarreta

Laresgoiti

, et al. Recycling of the products obtained in the pyrolysis of fibre‐glass polyester SMC. Chem Technol Biotechnol 1997; 69(2), 187–192. DOI: 10.1002/(SICI)1097-4660(199706)69:2%3C187::AID-JCTB710%3E3.0.CO;2-T

49.

Sokoli

Simonsen

Nielsen

, et al. Conversion of the matrix in glass fiber reinforced composites into a high heating value oil and other valuable feedstocks. Fuel Process Technol 2016; 149, 29–39. DOI: 10.1016/j.fuproc.2016.03.013

50.

Cunliffe

Jones

Williams

. Pyrolysis of composite plastic waste. Environ Technol 2003; 3330, 37–41. DOI: 10.1080/09593330309385599

51.

Onwudili

Miskolczi

Nagy

, et al. Recovery of glass fibre and carbon fibres from reinforced thermosets by batch pyrolysis and investigation of fibre re-using as reinforcement in LDPE matrix. Composites Part B 2016; 91, 154–161. DOI: 10.1016/j.compositesb.2016.01.055

52.

Kouparitsas

Kartalis

Varelidis

, et al. Recycling of the fibrous fraction of reinforced thermoset composites. Polym Compos 2002; 23(4), 682–689.

53.

Torres

De Marco

Caballero

, et al. Recycling of the solid residue obtained from the pyrolysis of fiberglass polyester sheet molding compound. Adv Polym Technol 2009; 28(2), 141–149. DOI: 10.1002/adv.20150