Article pubs.acs.org/est
Simultaneous Recovery of Benzene-Rich Oil and Metals by Steam Pyrolysis of Metal-Poly(ethylene terephthalate) Composite Waste Shogo Kumagai,†,‡ Guido Grause,† Tomohito Kameda,† and Toshiaki Yoshioka†,* †
Graduate School of Environmental Studies, Tohoku University, 6-6-07 Aoba, Aramaki-aza, Aoba-ku, Sendai, Miyagi 980-8579, Japan Japan Society for the Promotion of Science, 8 Ichibancho, Chiyoda-ku, Tokyo 102-8472, Japan
‡
S Supporting Information *
ABSTRACT: The possibility of simultaneous recovery of benzene and metals from the hydrolysis of poly(ethylene terephthalate) (PET)-based materials such as X-ray films, magnetic tape, and prepaid cards under a steam atmosphere at a temperature of 450 °C was evaluated. The hydrolysis resulted in metal-containing carbonaceous residue and volatile terephthalic acid (TPA). The effects of metals and additives on the recovery process were also investigated. All metals were quantitatively recovered, and silver, maghemite (γ-Fe2O3), and anatase (TiO2) were recovered without any changes in their crystal structures or compositions. In a second step, TPA was decarboxylized in the presence of calcium oxide (CaO) at 700 °C, producing benzene with an average yield of 34% and purity of 76%. Maghemite (γ-Fe2O3) incorporated in magnetic tape and prepaid cards could decarboxylate TPA. Aluminum present in the prepaid cards produced hydrogen by the reaction with steam. However, the presence of metals had no adverse influence on the recovery of benzene-rich oil in the presence of CaO. Therefore, this method can be applied to PET-based materials containing inorganic substances, which cannot be recycled effectively otherwise.
1. INTRODUCTION Over the decades, countless data storage systems have been developed, including photographic film, video tapes, CDs, and DVDs, for example, indicating the importance of data storage in our society. Despite differences in appearance, they share several commonalities: a plastic carrier and a metal containing layer for the information storage. One of the materials frequently used as a carrier is poly(ethylene terephthalate) (PET). Metals incorporated in the organic matrix include silver, iron, titanium, and so forth. These metals are often of economic interest, ignoring the possibility of recovery of organic materials. For example, silver from X-ray films is frequently recovered by incineration, oxidation, and subsequent electrolysis or by stripping the silver-containing gelatin layer by various solutions.1−3 These methods are aimed at recovering silver, while PET is degraded by oxidation. However, the amount of plastic in composites is generally larger than metals, and the majority of plastics are prepared from crude oil. Therefore, it is necessary to consider the organic matrix as a carbon-rich resource that can be converted into fuel and monomeric derivatives. Over a period of time, a wide variety of PET recycling methods have been investigated. The types of methods can be grouped into two categories: mechanical and feedstock recycling. It is well-known that PET bottles are made of high-purity PET resin, allowing direct mechanical recycling. In © 2014 American Chemical Society
contrast, PET fractions of lower quality can be converted into monomers by solvolysis such as hydrolysis,4−6 methanolysis,7,8 and glycolysis.9,10 The advantage of solvolysis is the recovery of monomers at a relatively low temperature. However, metals, fillers, and additives might have a negative impact on the process, making it difficult to use the solvent over time without regeneration. As opposed to mechanical recycling and solvolysis, pyrolysis in tube furnaces,11 fluidized-beds,12 and spouted-bed reactors13 is less sensitive to the presence of additives; however, valuable products are not obtained. Our previous work also indicated that pyrolysis has been shown to be ineffective for the recovery of precious metals from prepaid cards as this process converts PET to carbonaceous residue.12 Moreover, the formation of terephthalic acid (TPA) causes corrosion and clogging of the pipes in treating facilities14 because of its high sublimation point at around 400 °C. Calcium-based catalysts,15,16 iron oxyhydrate (FeOOH),17 and mixtures of calcium- and iron-based catalysts18 have been used in an attempt to suppress TPA generation. Hydrolysis in the presence of steam and CaO is a promising degradation method. Grause et al.19 hydrolyzed PET in a steam Received: Revised: Accepted: Published: 3430
November 12, 2013 February 13, 2014 February 14, 2014 February 14, 2014 dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
atmosphere using a fluidized bed reactor, resulting in a TPA yield of 72% and a small fraction of oligomers. Alternatively, TPA produced by the hydrolysis of PET was decarboxylated in the presence of Ca(OH)2, and benzene was obtained instead without producing sublimated substances.20−22 Furthermore, PET and TPA were decomposed using a tube reactor separated into two heating zones: one for PET hydrolysis and the other for TPA decarboxylation with CaO.23−26 As a result, benzene yields of 74% from PET and 84% from TPA were achieved, in both cases with a purity of 97 wt %. In the present work, a two-step process was developed by separating the steam hydrolysis of PET-metal composites from the TPA decarboxylation in the presence of CaO. At the firststep, PET was hydrolyzed by steam, and the TPA produced was locally separated from the first step decarboxylated by CaO, producing benzene-rich oil. The conglomeration of desired metals and CaO is prevented, and an additional separation process is avoided. In this work, the possibility of the simultaneous recovery of benzene-rich oil and metals was investigated. In addition, catalytic effects of the participating metals and organic additives are still unclear, therefore these were also investigated.
using gas chromatography equipped with a mass spectrometer (GC-MS) (Agilent technology, GC:HP6890, MS:HP5973). The content was determined by the weight difference before and after extraction. The gelatin layer on X-ray film was stripped using a 1 M solution of NaOH for 30 min at 80 °C, and the content was determined by the weight difference before and after stripping.1 CaO (particle size between 1.0 and 3.0 mm) was obtained from Okutama Industries Co., Ltd.26 Diethyl ether, ethanol, sodium hydroxide, concentrated sulfuric acid, concentrated nitric acid, hydrogen peroxide, standard solutions (silver, iron, titanium, and aluminum) for ICP, and naphthalene as an internal standard for GC were obtained from Kanto Chemical Co., Ltd. Carbon dioxide (CO2) standard gas with a purity of 99.9% used for GC analysis was obtained from GL-Science. Ion exchanged water was obtained using a water distillation apparatus (ADVANTEC, RFD240HA). 2.2. Pyrolysis Experiments. A diagram of the experimental system is shown in Figure 1. Details of equipment, procedure,
2. EXPERIMENTAL SECTION 2.1. Materials. Waste PET bottles and PET-metal composites, X-ray film, magnetic tape, and prepaid cards were cut into 2.8 × 2.8 mm2 pieces. Results of the elemental analysis and additive content are summarized in Table 1. Elemental Table 1. Elemental Analysis and Additive Content of PET Composites additive content/wt % bottle PET
X-ray film
PET resin (balance) Dioctyl pthalate gelatin layer metal/metal oxides
100.00
C H N Bromine Ag Fe Ti Al O (balance)
61.48 4.36 0.44 0.54 0.39 ± 0.04
61.93 4.34
magnetic tape
93.28 0.96 5.37 0.39 element analysis/wt
80.21 1.01
75.18 0.98
18.78 %
23.84
50.13 3.65 0.36
13.13 ± 0.20
33.73
32.79
prepaid cards
32.73
Figure 1. Experimental apparatus. (1) helium cylinder; (2) flow meter; (3) hydrolysis chamber; (4) decarboxylation chamber; (5) CaO bed; (6) perforated sample holder; (7) quartz tube reactor; (8) thermocouples; (9) electric furnaces; (10) coil heater; (11) steam generator; (12) water pump; (13) ion exchanged water reservoir; (14) ice trap; (15) liquid nitrogen trap; and (16) gas bag.
and analytical methods have been published elsewhere.25 Briefly, the sample holder connected to the top of the reactor was replaced with a perforated sample holder inside the quartz tube. The upper part of the furnace was assigned to the hydrolysis of PET and the lower part was filled with 16 g of CaO for the decarboxylation of TPA. Steam was produced by feeding ion-exchanged water into an electrically heated quartz tube. The perforated sample holder was filled with 0.5 g of sample and kept outside the heating zone until reaction conditions were steady. Before the sample holder was lowered into the reactor, CaO was calcined at 900 °C for 1 h under a constant helium flow of 35 mL min−1 in order to decompose potential calcium carbonate (CaCO3) at the bed material surface. After calcination, the temperatures of the hydrolysis chamber and decarboxylation chamber were set to 450 and 700 °C, respectively. When constant temperature was achieved, the steam concentration was adjusted to 88 vol % (35 mL min−1 helium, 257 mL min−1 steam at 25 °C). When constant carrier gas flow was achieved, the sample was lowered into the hydrolysis chamber. Gaseous TPA was carried by steam flow into the decarboxylation chamber where it reacted with CaO, while the metal-containing carbonaceous residue remained in
49.25 3.58 0.38
7.91 ± 0.13 7.29 ± 0.38 0.37 ± 0.01 31.22
composition (carbon, hydrogen, nitrogen, and bromine) of the samples was determined using a MICRO CORDER JM10 and a YANAKO MT-6 (both J-SCIENCE GROUP). The initial metal composition of the samples was determined qualitatively and quantitatively by inductively coupled plasma atomic emission spectrometry (ICP-AES) (Thermo Fisher Scientific Inc., iCAP 6500 Duo) following the decomposition of the organic sample matrix (see section 2.3). ICP analysis was determined for 3 consecutive runs for each sample, and standard deviations of the initial metal contents are included in Table 1. Oxygen was used as balance. In order to determine the additive content, Soxhlet extraction was carried out using diethyl ether for 6 h. Extracted compounds were identified 3431
dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
determination of gaseous products was carried out by the internal standard method using CO2 (5 mL) as the standard substance. Liquid products were identified by GC−MS (Column: InertCap5MS/Sil, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min)). Quantitative analyses of the liquid products were carried out by GC−FID (GL Science GC390, InertCap5MS/Sil, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min)) using naphthalene as the internal standard. 2.6. TG Analysis in Steam Atmosphere. In order to investigate the degradation behavior of PET-metal composites in the steam atmosphere, thermogravimetric analysis (TGA) was performed in a temperature range between 200 and 500 °C under a 75 vol% steam flow (balanced by helium) using a Seiko TG/DTA6200 controlled by a Seiko Exstar 6000 System. The TGA was quenched for 10 min with a helium flow of 65 mL/ min after placing 10 mg of the sample on the TG balance and the furnace was heated to 180 °C. Subsequently, 225 mL/min of steam was introduced into the furnace with 10 mL/min of He. When constant gas flow was achieved, the temperature was increased with a heating rate of 5 K/min. Additional experiments were carried out with a mixture of TPA and γFe2O3 (weight ratio TPA: γ-Fe2O3 = 79:21) in order to investigate the interaction between TPA and γ-Fe2O3. 2.7. Definitions. Figure 2 shows the reaction scheme of the degradation of PET-metal composites in the presence of CaO.
the sample holder. Liquid and gaseous products, as well as CaO were subsequently gathered and analyzed.25 Experiments in the absence of CaO were similarly carried out. However, deposits of sublimating substances at the outlet of the reactor tube were washed with ethanol, and benzoic acid (BA) was dissolved. The remaining residue was dissolved in 1 M NaOH and recrystallized using HCl. The product was filtered, washed with deionized water, dried, balanced, and identified as TPA by Fourier transform infrared spectroscopy (FT-IR) (Thermo-Fisher Scientific, Nicolet 6700, attenuated total reflection (ATR) method). Experiments were repeated at least 2 times to ensure the reliability of the results. No significant discrepancies were observed for each run under the same conditions. 2.3. Characterization of Metal Containing Residue. Solid products in the sample holder were recovered and balanced. Part of the recovered sample was analyzed using Xray diffraction (XRD) (Rigaku, Powder Diffractometer RINT2200VHF+/PC, between 3° and 90° using the Cu−Kα line), while another part was analyzed by a scanning electron microscope equipped with energy dispersive X-ray spectroscopy (SEM-EDX) (HITACHI, S-4800). To determine the metal content, part of the sample was dissolved in 20 mL concentrated sulfuric acid at 150 °C using a 100 mL three-neck flask equipped with a reflux condenser. When white fume was observed, 20 mL of concentrated nitric acid was added dropwise to the solution. After the organic matrix was completely decomposed, the solution was cooled down to room temperature, and 20 mL of hydrogen peroxide was added. The solution was diluted to 500 mL by adding ionexchanged water, and analyzed by ICP-AES. However, TiO2 dissolution was incomplete and therefore, collected and dried in vacuo at 40 °C for 1 day and subsequently weighed and analyzed by ICP-AES. 2.4. Quantitative Determination of CaCO3. Following the pyrolysis experiments, CaO from the upper part of the fixed bed (3 to 5 g from the top) covered by carbonaceous residue, and from the lower part (remaining 11 to 13 g) were separated and balanced. About 0.25 g of each CaO bed material fraction was placed in a 100 mL three-neck flask and independently analyzed. After replacing air with helium for 30 min, the flask was connected to a gas bag and 5 mL of hydrochloric acid was injected by a syringe through a septum attached to the flask. The CO2 released during the dissolution of the CaO bed material was driven by the helium flow into the gas bag and quantified by gas chromatography using a methanizer and a flame ionization detector (GC-FID). 2.5. Analysis of Organic Products. Gaseous products were identified by gas chromatography mass spectrometry (GC−MS) (Agilent technology, GC: HP6890, CP-PORA BOND Q, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min); MS: HP5973, 70 eV, 30−500 Da) using the MS library Wiley 275. Standard Chemistation (Hewlett-Packard G1017DA Revision D.01.02J) was used to operate the GC−MS. Quantitative analysis was carried out using GC equipped with a thermal conductivity detector (GC−TCD) (GL Science, GC323, active carbon (30−60 mesh size), 50 °C (3 min) → 10 °C min−1 → 150 °C (5 min)) and a flame ionization detector (GC−FID) (GL Science, GC4000, CP-PORA BOND Q, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min)). Carbon monoxide and CO2 were converted to methane by a methanizer (GL Science, MT 221) located between the column end and the FID detector. The quantitative
Figure 2. Benzene and metal recovery from PET-metal composites in the presence of CaO.
The composites were hydrolyzed in steam atmosphere, resulting in gaseous TPA, and metals were concentrated in solid residue. The carboxyl group of TPA is decarboxylated in the presence of CaO,25 resulting in benzene and CaCO3. CaO is regenerated by the decarbonation of CaCO3. Liquid and gaseous products were standardized based on the weight of PET. The weight fraction of the decomposition products was defined as follows: wt fract [wt%] =
prod wt [g] × 100% PET wt [g]
(1)
The complete conversion of PET (192 g/mol per repeating unit) would result in 41 wt % benzene (78 g/mol); hence, the benzene yield was defined as follows: benz yld [%] =
benz wt fract [wt%] × 100% 41 [wt%]
(2)
The benzene purity in the liquid fraction was defined as follows: 3432
dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
Table 2. Steam Pyrolysis Products of PET-Metal Composites in the Presence and Absence of CaOa experiment materials
B1
B2
F1
bottle PET
F2 X-ray film
CaO gases hydrogen methane ethane ethylene propylene propyne propadiene butene carbon monoxide carbon dioxide others liquids benzene benzene derivatives toluene xylene ethylbenzene styrene benzonitrile naphthalene and naphthalene derivatives naphthalene methylnaphthalene biphenyl and biphenyl derivatives biphenyl methylbiphenyl other cyclic compounds indene diphenylmethane diphenylethylene fluorene phenylstyrene phenanthrene anthracene phenylnaphthalene terphenyl triphenylmethane pyrene oxygen containing compounds phenol acetophenone methylacetophenone cyanoacetophenone toluicacid ethylbenzoic acid naphtoic acid diacetylbenzene benzaldehyde phenylbenzaldehyde methylbenzoate cyclopentanone fluorenone others sublimating substances benzoic acid terephthalic acid
T1
T2
magnetic tape
CaO
C1
C2
prepaid cards
CaO
CaO
20.8 − 2.6 0.1 0.6 0.1 + + 0.1 7.7 9.7 + 2.6 0.3 + + − + + − 0.1 + 0.1 0.1 0.1 + 0.2 + 0.1 − + − +
41.0 0.6 3.2 + 0.2 0.1 − − + 0.1 36.8 − 18.7 13.6 2.0 1.2 + + 0.8 − 0.2 0.2 + 2.5 2.4 0.1 0.3 − + + − 0.1 +
21.2 − 2.2 0.1 1.3 0.1 + + 0.1 5.4 12.0 − 2.8 0.6 0.4 0.1 + 0.1 0.2 + 0.1 + 0.1 0.2 0.2 + 0.2 + 0.1 + + + +
39.5 0.7 3.3 + 0.5 0.1 − − + 0.1 34.7 − 18.0 13.0 2.0 1.1 + 0.1 0.8 − 0.2 0.2 + 2.4 2.4 0.1 0.4 − 0.1 + − 0.1 +
19.8 − 2.2 0.1 1.1 0.3 + + 0.1 3.9 12.2 − 1.9 0.4 + − − − + − + + − 0.3 0.3 + 0.2 + + + 0.1 − +
48.3 0.3 2.5 + 0.6 0.2 − − 0.1 0.1 44.5 − 17.9 14.3 1.0 0.4 − − 0.5 − 0.1 0.1 − 2.1 2.1 0.1 0.3 − 0.1 − − − +
21.4 0.2 3.8 0.1 1.2 0.1 − + 0.1 5.4 10.3 − 2.6 0.4 0.4 0.2 − 0.1 0.1 − + + − 0.4 0.4 0.1 0.1 + + + + + +
53.4 0.7 3.9 0.1 0.8 0.1 − − 0.1 0.3 47.4 − 18.5 14.0 1.5 0.9 + + 0.6 − 0.2 0.2 − 2.3 2.3 + 0.3 − 0.1 − − − +
− + − − 0.8 − 0.5 0.1 − 0.1 − + 0.1 0.1 − − − + 1.1 21.1 6.6 14.5
− 0.1 + 0.1 − − − − − − − − − − − − − 0.1 − − −
− 0.1 − + 0.8 0.1 0.5 0.1 + + + 0.1 − − + + − + 0.5 21.8 6.9 14.9
+ 0.1 + 0.1 − − − − − − − − − − − − − − 0.1 − − −
− + − − 0.8 + 0.4 + − + − − + 0.3 − − − + 0.2 21.7 11.7 9.9
− 0.1 − 0.1 + + − − − − − − − − − − − − 0.1 − − −
+ − − − 0.6 − 0.3 − − + − 0.1 + − − − 0.2 + 0.7 20.3 7.9 12.5
+ 0.1 − 0.1 0.1 0.1 − − − − − − − − − − − − 0.2 − − −
3433
dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
Table 2. continued experiment
B1
materials
B2
F1
bottle PET
F2 X-ray film
CaO carbonaceous residue (in sample holder) total identif ied products/wt % not identif ied carbon/c % benzene yield/% benzene purity/wt % a
12.5 57.0 41.6 0.8 12.9
T1
magnetic tape
CaO
14.6 73.7 29.0 33.4 72.6
12.3 60.4 36.7 1.5 21.3
10.9 71.0 30.9 31.9 72.1
T2
C1
prepaid cards
CaO 14.0 55.7 42.0 0.9 18.8
13.5 77.6 30.3 35.2 79.9
C2
CaO 12.4 56.7 43.2 0.9 14.8
13.2 85.1 27.1 34.5 76.0
−, Not detected; +,
Simultaneous Recovery of Benzene-Rich Oil and Metals by Steam Pyrolysis of Metal-Poly(ethylene terephthalate) Composite Waste Shogo Kumagai,†,‡ Guido Grause,† Tomohito Kameda,† and Toshiaki Yoshioka†,* †
Graduate School of Environmental Studies, Tohoku University, 6-6-07 Aoba, Aramaki-aza, Aoba-ku, Sendai, Miyagi 980-8579, Japan Japan Society for the Promotion of Science, 8 Ichibancho, Chiyoda-ku, Tokyo 102-8472, Japan
‡
S Supporting Information *
ABSTRACT: The possibility of simultaneous recovery of benzene and metals from the hydrolysis of poly(ethylene terephthalate) (PET)-based materials such as X-ray films, magnetic tape, and prepaid cards under a steam atmosphere at a temperature of 450 °C was evaluated. The hydrolysis resulted in metal-containing carbonaceous residue and volatile terephthalic acid (TPA). The effects of metals and additives on the recovery process were also investigated. All metals were quantitatively recovered, and silver, maghemite (γ-Fe2O3), and anatase (TiO2) were recovered without any changes in their crystal structures or compositions. In a second step, TPA was decarboxylized in the presence of calcium oxide (CaO) at 700 °C, producing benzene with an average yield of 34% and purity of 76%. Maghemite (γ-Fe2O3) incorporated in magnetic tape and prepaid cards could decarboxylate TPA. Aluminum present in the prepaid cards produced hydrogen by the reaction with steam. However, the presence of metals had no adverse influence on the recovery of benzene-rich oil in the presence of CaO. Therefore, this method can be applied to PET-based materials containing inorganic substances, which cannot be recycled effectively otherwise.
1. INTRODUCTION Over the decades, countless data storage systems have been developed, including photographic film, video tapes, CDs, and DVDs, for example, indicating the importance of data storage in our society. Despite differences in appearance, they share several commonalities: a plastic carrier and a metal containing layer for the information storage. One of the materials frequently used as a carrier is poly(ethylene terephthalate) (PET). Metals incorporated in the organic matrix include silver, iron, titanium, and so forth. These metals are often of economic interest, ignoring the possibility of recovery of organic materials. For example, silver from X-ray films is frequently recovered by incineration, oxidation, and subsequent electrolysis or by stripping the silver-containing gelatin layer by various solutions.1−3 These methods are aimed at recovering silver, while PET is degraded by oxidation. However, the amount of plastic in composites is generally larger than metals, and the majority of plastics are prepared from crude oil. Therefore, it is necessary to consider the organic matrix as a carbon-rich resource that can be converted into fuel and monomeric derivatives. Over a period of time, a wide variety of PET recycling methods have been investigated. The types of methods can be grouped into two categories: mechanical and feedstock recycling. It is well-known that PET bottles are made of high-purity PET resin, allowing direct mechanical recycling. In © 2014 American Chemical Society
contrast, PET fractions of lower quality can be converted into monomers by solvolysis such as hydrolysis,4−6 methanolysis,7,8 and glycolysis.9,10 The advantage of solvolysis is the recovery of monomers at a relatively low temperature. However, metals, fillers, and additives might have a negative impact on the process, making it difficult to use the solvent over time without regeneration. As opposed to mechanical recycling and solvolysis, pyrolysis in tube furnaces,11 fluidized-beds,12 and spouted-bed reactors13 is less sensitive to the presence of additives; however, valuable products are not obtained. Our previous work also indicated that pyrolysis has been shown to be ineffective for the recovery of precious metals from prepaid cards as this process converts PET to carbonaceous residue.12 Moreover, the formation of terephthalic acid (TPA) causes corrosion and clogging of the pipes in treating facilities14 because of its high sublimation point at around 400 °C. Calcium-based catalysts,15,16 iron oxyhydrate (FeOOH),17 and mixtures of calcium- and iron-based catalysts18 have been used in an attempt to suppress TPA generation. Hydrolysis in the presence of steam and CaO is a promising degradation method. Grause et al.19 hydrolyzed PET in a steam Received: Revised: Accepted: Published: 3430
November 12, 2013 February 13, 2014 February 14, 2014 February 14, 2014 dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
atmosphere using a fluidized bed reactor, resulting in a TPA yield of 72% and a small fraction of oligomers. Alternatively, TPA produced by the hydrolysis of PET was decarboxylated in the presence of Ca(OH)2, and benzene was obtained instead without producing sublimated substances.20−22 Furthermore, PET and TPA were decomposed using a tube reactor separated into two heating zones: one for PET hydrolysis and the other for TPA decarboxylation with CaO.23−26 As a result, benzene yields of 74% from PET and 84% from TPA were achieved, in both cases with a purity of 97 wt %. In the present work, a two-step process was developed by separating the steam hydrolysis of PET-metal composites from the TPA decarboxylation in the presence of CaO. At the firststep, PET was hydrolyzed by steam, and the TPA produced was locally separated from the first step decarboxylated by CaO, producing benzene-rich oil. The conglomeration of desired metals and CaO is prevented, and an additional separation process is avoided. In this work, the possibility of the simultaneous recovery of benzene-rich oil and metals was investigated. In addition, catalytic effects of the participating metals and organic additives are still unclear, therefore these were also investigated.
using gas chromatography equipped with a mass spectrometer (GC-MS) (Agilent technology, GC:HP6890, MS:HP5973). The content was determined by the weight difference before and after extraction. The gelatin layer on X-ray film was stripped using a 1 M solution of NaOH for 30 min at 80 °C, and the content was determined by the weight difference before and after stripping.1 CaO (particle size between 1.0 and 3.0 mm) was obtained from Okutama Industries Co., Ltd.26 Diethyl ether, ethanol, sodium hydroxide, concentrated sulfuric acid, concentrated nitric acid, hydrogen peroxide, standard solutions (silver, iron, titanium, and aluminum) for ICP, and naphthalene as an internal standard for GC were obtained from Kanto Chemical Co., Ltd. Carbon dioxide (CO2) standard gas with a purity of 99.9% used for GC analysis was obtained from GL-Science. Ion exchanged water was obtained using a water distillation apparatus (ADVANTEC, RFD240HA). 2.2. Pyrolysis Experiments. A diagram of the experimental system is shown in Figure 1. Details of equipment, procedure,
2. EXPERIMENTAL SECTION 2.1. Materials. Waste PET bottles and PET-metal composites, X-ray film, magnetic tape, and prepaid cards were cut into 2.8 × 2.8 mm2 pieces. Results of the elemental analysis and additive content are summarized in Table 1. Elemental Table 1. Elemental Analysis and Additive Content of PET Composites additive content/wt % bottle PET
X-ray film
PET resin (balance) Dioctyl pthalate gelatin layer metal/metal oxides
100.00
C H N Bromine Ag Fe Ti Al O (balance)
61.48 4.36 0.44 0.54 0.39 ± 0.04
61.93 4.34
magnetic tape
93.28 0.96 5.37 0.39 element analysis/wt
80.21 1.01
75.18 0.98
18.78 %
23.84
50.13 3.65 0.36
13.13 ± 0.20
33.73
32.79
prepaid cards
32.73
Figure 1. Experimental apparatus. (1) helium cylinder; (2) flow meter; (3) hydrolysis chamber; (4) decarboxylation chamber; (5) CaO bed; (6) perforated sample holder; (7) quartz tube reactor; (8) thermocouples; (9) electric furnaces; (10) coil heater; (11) steam generator; (12) water pump; (13) ion exchanged water reservoir; (14) ice trap; (15) liquid nitrogen trap; and (16) gas bag.
and analytical methods have been published elsewhere.25 Briefly, the sample holder connected to the top of the reactor was replaced with a perforated sample holder inside the quartz tube. The upper part of the furnace was assigned to the hydrolysis of PET and the lower part was filled with 16 g of CaO for the decarboxylation of TPA. Steam was produced by feeding ion-exchanged water into an electrically heated quartz tube. The perforated sample holder was filled with 0.5 g of sample and kept outside the heating zone until reaction conditions were steady. Before the sample holder was lowered into the reactor, CaO was calcined at 900 °C for 1 h under a constant helium flow of 35 mL min−1 in order to decompose potential calcium carbonate (CaCO3) at the bed material surface. After calcination, the temperatures of the hydrolysis chamber and decarboxylation chamber were set to 450 and 700 °C, respectively. When constant temperature was achieved, the steam concentration was adjusted to 88 vol % (35 mL min−1 helium, 257 mL min−1 steam at 25 °C). When constant carrier gas flow was achieved, the sample was lowered into the hydrolysis chamber. Gaseous TPA was carried by steam flow into the decarboxylation chamber where it reacted with CaO, while the metal-containing carbonaceous residue remained in
49.25 3.58 0.38
7.91 ± 0.13 7.29 ± 0.38 0.37 ± 0.01 31.22
composition (carbon, hydrogen, nitrogen, and bromine) of the samples was determined using a MICRO CORDER JM10 and a YANAKO MT-6 (both J-SCIENCE GROUP). The initial metal composition of the samples was determined qualitatively and quantitatively by inductively coupled plasma atomic emission spectrometry (ICP-AES) (Thermo Fisher Scientific Inc., iCAP 6500 Duo) following the decomposition of the organic sample matrix (see section 2.3). ICP analysis was determined for 3 consecutive runs for each sample, and standard deviations of the initial metal contents are included in Table 1. Oxygen was used as balance. In order to determine the additive content, Soxhlet extraction was carried out using diethyl ether for 6 h. Extracted compounds were identified 3431
dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
determination of gaseous products was carried out by the internal standard method using CO2 (5 mL) as the standard substance. Liquid products were identified by GC−MS (Column: InertCap5MS/Sil, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min)). Quantitative analyses of the liquid products were carried out by GC−FID (GL Science GC390, InertCap5MS/Sil, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min)) using naphthalene as the internal standard. 2.6. TG Analysis in Steam Atmosphere. In order to investigate the degradation behavior of PET-metal composites in the steam atmosphere, thermogravimetric analysis (TGA) was performed in a temperature range between 200 and 500 °C under a 75 vol% steam flow (balanced by helium) using a Seiko TG/DTA6200 controlled by a Seiko Exstar 6000 System. The TGA was quenched for 10 min with a helium flow of 65 mL/ min after placing 10 mg of the sample on the TG balance and the furnace was heated to 180 °C. Subsequently, 225 mL/min of steam was introduced into the furnace with 10 mL/min of He. When constant gas flow was achieved, the temperature was increased with a heating rate of 5 K/min. Additional experiments were carried out with a mixture of TPA and γFe2O3 (weight ratio TPA: γ-Fe2O3 = 79:21) in order to investigate the interaction between TPA and γ-Fe2O3. 2.7. Definitions. Figure 2 shows the reaction scheme of the degradation of PET-metal composites in the presence of CaO.
the sample holder. Liquid and gaseous products, as well as CaO were subsequently gathered and analyzed.25 Experiments in the absence of CaO were similarly carried out. However, deposits of sublimating substances at the outlet of the reactor tube were washed with ethanol, and benzoic acid (BA) was dissolved. The remaining residue was dissolved in 1 M NaOH and recrystallized using HCl. The product was filtered, washed with deionized water, dried, balanced, and identified as TPA by Fourier transform infrared spectroscopy (FT-IR) (Thermo-Fisher Scientific, Nicolet 6700, attenuated total reflection (ATR) method). Experiments were repeated at least 2 times to ensure the reliability of the results. No significant discrepancies were observed for each run under the same conditions. 2.3. Characterization of Metal Containing Residue. Solid products in the sample holder were recovered and balanced. Part of the recovered sample was analyzed using Xray diffraction (XRD) (Rigaku, Powder Diffractometer RINT2200VHF+/PC, between 3° and 90° using the Cu−Kα line), while another part was analyzed by a scanning electron microscope equipped with energy dispersive X-ray spectroscopy (SEM-EDX) (HITACHI, S-4800). To determine the metal content, part of the sample was dissolved in 20 mL concentrated sulfuric acid at 150 °C using a 100 mL three-neck flask equipped with a reflux condenser. When white fume was observed, 20 mL of concentrated nitric acid was added dropwise to the solution. After the organic matrix was completely decomposed, the solution was cooled down to room temperature, and 20 mL of hydrogen peroxide was added. The solution was diluted to 500 mL by adding ionexchanged water, and analyzed by ICP-AES. However, TiO2 dissolution was incomplete and therefore, collected and dried in vacuo at 40 °C for 1 day and subsequently weighed and analyzed by ICP-AES. 2.4. Quantitative Determination of CaCO3. Following the pyrolysis experiments, CaO from the upper part of the fixed bed (3 to 5 g from the top) covered by carbonaceous residue, and from the lower part (remaining 11 to 13 g) were separated and balanced. About 0.25 g of each CaO bed material fraction was placed in a 100 mL three-neck flask and independently analyzed. After replacing air with helium for 30 min, the flask was connected to a gas bag and 5 mL of hydrochloric acid was injected by a syringe through a septum attached to the flask. The CO2 released during the dissolution of the CaO bed material was driven by the helium flow into the gas bag and quantified by gas chromatography using a methanizer and a flame ionization detector (GC-FID). 2.5. Analysis of Organic Products. Gaseous products were identified by gas chromatography mass spectrometry (GC−MS) (Agilent technology, GC: HP6890, CP-PORA BOND Q, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min); MS: HP5973, 70 eV, 30−500 Da) using the MS library Wiley 275. Standard Chemistation (Hewlett-Packard G1017DA Revision D.01.02J) was used to operate the GC−MS. Quantitative analysis was carried out using GC equipped with a thermal conductivity detector (GC−TCD) (GL Science, GC323, active carbon (30−60 mesh size), 50 °C (3 min) → 10 °C min−1 → 150 °C (5 min)) and a flame ionization detector (GC−FID) (GL Science, GC4000, CP-PORA BOND Q, 50 °C (5 min) → 5 °C min−1 → 320 °C (10 min)). Carbon monoxide and CO2 were converted to methane by a methanizer (GL Science, MT 221) located between the column end and the FID detector. The quantitative
Figure 2. Benzene and metal recovery from PET-metal composites in the presence of CaO.
The composites were hydrolyzed in steam atmosphere, resulting in gaseous TPA, and metals were concentrated in solid residue. The carboxyl group of TPA is decarboxylated in the presence of CaO,25 resulting in benzene and CaCO3. CaO is regenerated by the decarbonation of CaCO3. Liquid and gaseous products were standardized based on the weight of PET. The weight fraction of the decomposition products was defined as follows: wt fract [wt%] =
prod wt [g] × 100% PET wt [g]
(1)
The complete conversion of PET (192 g/mol per repeating unit) would result in 41 wt % benzene (78 g/mol); hence, the benzene yield was defined as follows: benz yld [%] =
benz wt fract [wt%] × 100% 41 [wt%]
(2)
The benzene purity in the liquid fraction was defined as follows: 3432
dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
Table 2. Steam Pyrolysis Products of PET-Metal Composites in the Presence and Absence of CaOa experiment materials
B1
B2
F1
bottle PET
F2 X-ray film
CaO gases hydrogen methane ethane ethylene propylene propyne propadiene butene carbon monoxide carbon dioxide others liquids benzene benzene derivatives toluene xylene ethylbenzene styrene benzonitrile naphthalene and naphthalene derivatives naphthalene methylnaphthalene biphenyl and biphenyl derivatives biphenyl methylbiphenyl other cyclic compounds indene diphenylmethane diphenylethylene fluorene phenylstyrene phenanthrene anthracene phenylnaphthalene terphenyl triphenylmethane pyrene oxygen containing compounds phenol acetophenone methylacetophenone cyanoacetophenone toluicacid ethylbenzoic acid naphtoic acid diacetylbenzene benzaldehyde phenylbenzaldehyde methylbenzoate cyclopentanone fluorenone others sublimating substances benzoic acid terephthalic acid
T1
T2
magnetic tape
CaO
C1
C2
prepaid cards
CaO
CaO
20.8 − 2.6 0.1 0.6 0.1 + + 0.1 7.7 9.7 + 2.6 0.3 + + − + + − 0.1 + 0.1 0.1 0.1 + 0.2 + 0.1 − + − +
41.0 0.6 3.2 + 0.2 0.1 − − + 0.1 36.8 − 18.7 13.6 2.0 1.2 + + 0.8 − 0.2 0.2 + 2.5 2.4 0.1 0.3 − + + − 0.1 +
21.2 − 2.2 0.1 1.3 0.1 + + 0.1 5.4 12.0 − 2.8 0.6 0.4 0.1 + 0.1 0.2 + 0.1 + 0.1 0.2 0.2 + 0.2 + 0.1 + + + +
39.5 0.7 3.3 + 0.5 0.1 − − + 0.1 34.7 − 18.0 13.0 2.0 1.1 + 0.1 0.8 − 0.2 0.2 + 2.4 2.4 0.1 0.4 − 0.1 + − 0.1 +
19.8 − 2.2 0.1 1.1 0.3 + + 0.1 3.9 12.2 − 1.9 0.4 + − − − + − + + − 0.3 0.3 + 0.2 + + + 0.1 − +
48.3 0.3 2.5 + 0.6 0.2 − − 0.1 0.1 44.5 − 17.9 14.3 1.0 0.4 − − 0.5 − 0.1 0.1 − 2.1 2.1 0.1 0.3 − 0.1 − − − +
21.4 0.2 3.8 0.1 1.2 0.1 − + 0.1 5.4 10.3 − 2.6 0.4 0.4 0.2 − 0.1 0.1 − + + − 0.4 0.4 0.1 0.1 + + + + + +
53.4 0.7 3.9 0.1 0.8 0.1 − − 0.1 0.3 47.4 − 18.5 14.0 1.5 0.9 + + 0.6 − 0.2 0.2 − 2.3 2.3 + 0.3 − 0.1 − − − +
− + − − 0.8 − 0.5 0.1 − 0.1 − + 0.1 0.1 − − − + 1.1 21.1 6.6 14.5
− 0.1 + 0.1 − − − − − − − − − − − − − 0.1 − − −
− 0.1 − + 0.8 0.1 0.5 0.1 + + + 0.1 − − + + − + 0.5 21.8 6.9 14.9
+ 0.1 + 0.1 − − − − − − − − − − − − − − 0.1 − − −
− + − − 0.8 + 0.4 + − + − − + 0.3 − − − + 0.2 21.7 11.7 9.9
− 0.1 − 0.1 + + − − − − − − − − − − − − 0.1 − − −
+ − − − 0.6 − 0.3 − − + − 0.1 + − − − 0.2 + 0.7 20.3 7.9 12.5
+ 0.1 − 0.1 0.1 0.1 − − − − − − − − − − − − 0.2 − − −
3433
dx.doi.org/10.1021/es405047j | Environ. Sci. Technol. 2014, 48, 3430−3437
Environmental Science & Technology
Article
Table 2. continued experiment
B1
materials
B2
F1
bottle PET
F2 X-ray film
CaO carbonaceous residue (in sample holder) total identif ied products/wt % not identif ied carbon/c % benzene yield/% benzene purity/wt % a
12.5 57.0 41.6 0.8 12.9
T1
magnetic tape
CaO
14.6 73.7 29.0 33.4 72.6
12.3 60.4 36.7 1.5 21.3
10.9 71.0 30.9 31.9 72.1
T2
C1
prepaid cards
CaO 14.0 55.7 42.0 0.9 18.8
13.5 77.6 30.3 35.2 79.9
C2
CaO 12.4 56.7 43.2 0.9 14.8
13.2 85.1 27.1 34.5 76.0
−, Not detected; +,
