Pyrolysis–gasification of plastics, mixed plastics and real-world plastic waste with and without Ni–Mg–Al catalyst Chunfei Wu, Paul T. Williams * Energy & Resources Research Institute, The University of Leeds, Leeds LS2 9JT, UK article info Article history: Received 27 July 2009 Received in revised form 29 April 2010 Accepted 20 May 2010 Available online 2 June 2010 Keywords: Polypropylene Polystyrene Polyethylene Waste plastic Gasification abstract Polypropylene, polystyrene, high density polyethylene and their mixtures and real-world plastic waste were investigated for the production of hydrogen in a two-stage pyrolysis–gasification reactor. The experiments were carried out at gasification temperatures of 800 or 850 °C with or without a Ni–Mg– Al catalyst. The influence of plastic type on the product distribution and hydrogen production in relation to process conditions were investigated. The reacted Ni–Mg–Al catalysts were analyzed by temperature- programmed oxidation and scanning electron microscopy. The results showed that lower gas yield (11.2 wt.% related to the mass of plastic) was obtained for the non-catalytic non-steam pyrolysis–gasification of polystyrene at the gasification temperature of 800 °C, compared with the polypropylene (59.6 wt.%) and high density polyethylene (53.5 wt.%) and waste plastic (45.5 wt.%). In addition, the largest oil prod- uct was observed for the non-catalytic pyrolysis–gasification of polystyrene. The presence of the Ni–Mg– Al catalyst greatly improved the steam pyrolysis–gasification of plastics for hydrogen production. The steam catalytic pyrolysis–gasification of polystyrene presented the lowest hydrogen production of 0.155 and 0.196 (g H 2 /g polystyrene) at the gasification temperatures of 800 and 850 °C, respectively. More coke was deposited on the catalyst for the pyrolysis–gasification of polypropylene and waste plastic compared with steam catalytic pyrolysis–gasification of polystyrene and high density polyethylene. Fil- amentous carbons were observed for the used Ni–Mg–Al catalysts from the pyrolysis–gasification of polypropylene, high density polyethylene, waste plastic and mixed plastics. However, the formation of filamentous carbons on the coked catalyst from the pyrolysis–gasification of polystyrene was low. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Increasing amounts of waste plastics are generated every year due to the increasing demand for the applications of plastics in the modern world [1]. Currently, the most common ways to treat waste plastics which arise through the municipal solid waste stream is via landfill and incineration. The landfilling of waste plas- tics means the energy content of the plastics is not recovered. Incineration, although recovering energy, results in the carbon con- tent in the plastics being mostly converted into CO 2 and released into the atmosphere. However, the recycling rate for the plastic fraction of municipal solid waste remains below 15%, representing a waste of a resource [1]. Feedstock recycling through the thermal degradation of waste plastics is a promising alternative for the management of waste plastics [2–6]. Gasification of plastics to generate hydrogen has re- cently attracted interest [7–9], due to the emphasis on a hydrogen energy future [10,11]. Because of the complexity of real-world waste plastics, processes that can accommodate mixtures of plas- tics to generate high value products are of interest. Gasification of plastics results in the production of a wide range of hydrocar- bons, however a two-stage pyrolysis–gasification system has been shown to produce high yield hydrogen from plastics [8,9]. The two- stage pyrolysis–gasification reaction system results in pyrolysis of the plastic waste followed by steam gasification of the product pyrolysis gases in the presence of a catalyst to produce hydrogen. The process is more controllable than gasification and is particu- larly suited to mixed plastic wastes, where any residues and dirt associated with the plastics remain in the pyrolysis unit. There are five main plastics which arise in European municipal solid waste which are polyethylene (as low and high density poly- ethylene), polypropylene, polystyrene polyethylene terephthalate and polyvinyl chloride. Polypropylene, polyethylene and polysty- rene are the main components of the plastics fraction in municipal solid waste [12]. In this paper, polypropylene, polystyrene, high density polyeth- ylene and mixtures of these plastics and also real-world plastic waste have been investigated using a two-stage pyrolysis–gasifica- tion reactor system. The purpose was to understand the production 0016-2361/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.fuel.2010.05.032 * Corresponding author. Tel.: +44 1133432504. E-mail address: [email protected](P.T. Williams). Fuel 89 (2010) 3022–3032 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel
Transcript
Fuel 89 (2010) 3022–3032
Contents lists available at ScienceDirect
Fuel
journal homepage: www.elsevier .com/locate / fuel
Pyrolysis–gasification of plastics, mixed plastics and real-world plastic wastewith and without Ni–Mg–Al catalyst
Chunfei Wu, Paul T. Williams *
Energy & Resources Research Institute, The University of Leeds, Leeds LS2 9JT, UK
a r t i c l e i n f o a b s t r a c t
Article history:Received 27 July 2009Received in revised form 29 April 2010Accepted 20 May 2010Available online 2 June 2010
Polypropylene, polystyrene, high density polyethylene and their mixtures and real-world plastic wastewere investigated for the production of hydrogen in a two-stage pyrolysis–gasification reactor. Theexperiments were carried out at gasification temperatures of 800 or 850 �C with or without a Ni–Mg–Al catalyst. The influence of plastic type on the product distribution and hydrogen production in relationto process conditions were investigated. The reacted Ni–Mg–Al catalysts were analyzed by temperature-programmed oxidation and scanning electron microscopy. The results showed that lower gas yield (11.2wt.% related to the mass of plastic) was obtained for the non-catalytic non-steam pyrolysis–gasificationof polystyrene at the gasification temperature of 800 �C, compared with the polypropylene (59.6 wt.%)and high density polyethylene (53.5 wt.%) and waste plastic (45.5 wt.%). In addition, the largest oil prod-uct was observed for the non-catalytic pyrolysis–gasification of polystyrene. The presence of the Ni–Mg–Al catalyst greatly improved the steam pyrolysis–gasification of plastics for hydrogen production. Thesteam catalytic pyrolysis–gasification of polystyrene presented the lowest hydrogen production of0.155 and 0.196 (g H2/g polystyrene) at the gasification temperatures of 800 and 850 �C, respectively.More coke was deposited on the catalyst for the pyrolysis–gasification of polypropylene and waste plasticcompared with steam catalytic pyrolysis–gasification of polystyrene and high density polyethylene. Fil-amentous carbons were observed for the used Ni–Mg–Al catalysts from the pyrolysis–gasification ofpolypropylene, high density polyethylene, waste plastic and mixed plastics. However, the formation offilamentous carbons on the coked catalyst from the pyrolysis–gasification of polystyrene was low.
� 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Increasing amounts of waste plastics are generated every yeardue to the increasing demand for the applications of plastics inthe modern world [1]. Currently, the most common ways to treatwaste plastics which arise through the municipal solid wastestream is via landfill and incineration. The landfilling of waste plas-tics means the energy content of the plastics is not recovered.Incineration, although recovering energy, results in the carbon con-tent in the plastics being mostly converted into CO2 and releasedinto the atmosphere. However, the recycling rate for the plasticfraction of municipal solid waste remains below 15%, representinga waste of a resource [1].
Feedstock recycling through the thermal degradation of wasteplastics is a promising alternative for the management of wasteplastics [2–6]. Gasification of plastics to generate hydrogen has re-cently attracted interest [7–9], due to the emphasis on a hydrogenenergy future [10,11]. Because of the complexity of real-world
ll rights reserved.
iams).
waste plastics, processes that can accommodate mixtures of plas-tics to generate high value products are of interest. Gasificationof plastics results in the production of a wide range of hydrocar-bons, however a two-stage pyrolysis–gasification system has beenshown to produce high yield hydrogen from plastics [8,9]. The two-stage pyrolysis–gasification reaction system results in pyrolysis ofthe plastic waste followed by steam gasification of the productpyrolysis gases in the presence of a catalyst to produce hydrogen.The process is more controllable than gasification and is particu-larly suited to mixed plastic wastes, where any residues and dirtassociated with the plastics remain in the pyrolysis unit.
There are five main plastics which arise in European municipalsolid waste which are polyethylene (as low and high density poly-ethylene), polypropylene, polystyrene polyethylene terephthalateand polyvinyl chloride. Polypropylene, polyethylene and polysty-rene are the main components of the plastics fraction in municipalsolid waste [12].
In this paper, polypropylene, polystyrene, high density polyeth-ylene and mixtures of these plastics and also real-world plasticwaste have been investigated using a two-stage pyrolysis–gasifica-tion reactor system. The purpose was to understand the production
C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032 3023
of hydrogen from different plastics by the pyrolysis–gasificationprocess with or without the presence of a Ni–Mg–Al catalyst.
2. Materials and methods
2.1. Materials
Polypropylene (PP) was obtained as 2 mm virgin polymer pel-lets provided by BP Chemicals UK. High density polyethylene(about 2 mm) (HDPE) and polystyrene (about 2 mm) (PS) were ob-tained from ACROS Organics UK. The mixtures of plastics consistedof 26.9 wt.% PP, 16.8 wt.% PS and 56.3 wt.% HDPE.
The waste plastic was real-world, post-consumer, municipal so-lid waste mixed plastic from Belgium collected and recycled byFost Plus, Belgium. The plastic waste had been flaked and sepa-rated into a low density fraction through air separation. The sam-ple size was approximately 5 mm sized flakes. The Fost Plus plasticconsisted of mainly high density polyethylene and polyethyleneterephthalate. The elemental analysis of waste plastics showedthat the C, H, O and N weight contents were 77.1, 11.5, 11.2 and0.2 wt.%, respectively, determined by using a CE InstrumentsCHNS-O analyzer. In addition, the proximate analysis resultsshowed that there was almost no moisture in the Fost Plus plasticand volatile, fixed carbon and ash component was about 96.3, 1.1and 2.6 wt.%, respectively.
Ni–Mg–Al catalysts were prepared using the rising pH tech-nique according to the method reported by Garcia et al. [13]. Theprecipitant 1 M NH4(OH) was added to 200 ml of an aqueous solu-tion containing Ni(NO3)�6H2O, Al(NO3)3�9H2O and Mg(NO3)2�6H2O.The precipitation was carried out at 40 �C with moderate stirringuntil the final pH (8.3) was obtained. The precipitates were filteredand washed with water (40 �C), followed by drying at 105 �C over-night, and then were calcined at 750 �C for 3 h. The initial Ni–Mg–Al molar ratio was 1:1:1. The Ni–Mg–Al (1:1:1) catalysts werecrushed and sieved to granules with a size range between 0.065and 0.212 mm.
2.2. Characterization of reacted catalysts
Temperature-programmed oxidation (TPO) of reacted catalystswas carried out using a Stanton-Redcroft thermogravimetric ana-lyzer (TGA) to determine the properties of the coked carbonsdeposited on the reacted catalysts. The differential thermo-gravim-etry (DTG) results from the TPO experiments are also discussed inthis paper. About 100 mg of the reacted catalyst was heated in anatmosphere of air at 15 �C min�1 to a final temperature of 800 �C,with a dwell time of 10 min.
High resolution scanning electron microscopy (SEM, LEO 1530)was used to characterize and examine the characteristics of thecarbon deposited on the coked catalysts.
2.3. Experimental system
The experimental system consisted of a two-stage pyrolysis–gasification reactor. The plastics were pyrolysed in the first reactor,where the temperature was 500 �C. The generated gaseous prod-ucts were then passed through to the second reactor, where thetemperature was maintained at 800 or 850 �C. 0.5 g of Ni–Mg–Alcatalyst was placed in the second reactor. When the experimentwas carried out without Ni–Mg–Al catalyst, 0.5 g sand was usedas the substitute for the catalyst. Nitrogen was used as the carriergas with a flow rate of 60 ml min�1. The procedure was to heat thesecond gasification reactor to the desired temperature, then heatthe first reactor to 500 �C with a heating rate of 40 �C min�1. Waterwas injected with a flow rate of 4.74 g h�1 into the second reactor
via a syringe pump, therefore passing steam through the catalystbed (gasification) together with the pyrolysed gases derived fromthe thermal degradation of the plastic from the first stage reactor.The reaction time was 30 min after the water was injected. A sche-matic diagram of the experimental system was shown in our pre-vious work [8].
The gaseous products after the gasification process were passedthrough two condensers, where any condensed products were col-lected. The non-condensed gases were collected with a 25 L Ted-lar™ gas sample bag. The reproducibility of the reaction systemwas tested thoroughly, and experiments were repeated to ensurethe reliability of research results.
The gases collected in the sample bag were analyzed by packedcolumn gas chromatography (GC). Hydrocarbons (C1–C4) were ana-lyzed using a Varian 3380 gas chromatograph with a flame ionisa-tion detector, with a 80–100 mesh Hysep column and nitrogencarrier gas. Permanent gases (H2, CO, O2, N2 and CO2) were ana-lyzed by a second Varian 3380 GC with two separate columns.Hydrogen, oxygen, carbon monoxide and nitrogen were analyzedon a 60–80 mesh molecular sieve column with argon carrier gas.Carbon dioxide was analyzed on a Hysep 80–100 mesh columnwith argon carrier gas.
3. Results and discussion
3.1. Mass balances and hydrogen production without catalyst
3.1.1. Mass balances without catalyst and without steamThe pyrolysis–gasification of PP, PS, HDPE and waste plastic
were carried out without the presence of a catalyst (sand wassubstituted for the Ni–Mg–Al catalyst) at the gasification tempera-ture of 800 �C. The product yields and mass balances of the exper-imental results are shown in Table 1.
Table 1 shows that about 59.6 wt.% of gas yield related to themass of PP was obtained, when no steam was introduced intothe gasification reactor at 800 �C. At the same conditions (no steamand no catalyst), 53.5 wt.% of gas yield was obtained for the pyro-lysis–gasification of HDPE, and 45.5 wt.% for the real-world, FostPlus waste plastic. However, only 11.6 wt.% of gas yield corre-sponding to the mass of PS was obtained for the non-catalytic,non-steam pyrolysis–gasification of PS at the temperature of800 �C. It is suggested that the gaseous products derived fromthe pyrolysis of PS might need higher reaction energy to be crackedcompared to the other plastics investigated here. Therefore, moreoil and less solid product were obtained for PS compared to PPand HDPE, when the plastics were pyrolysed at 500 �C and thengasified at 800 �C without steam and without catalyst.
Bagri and Williams [4,14] have investigated the pyrolysis ofpolyethylene (PE) and PS with a two-stage reactor system, involv-ing a first stage pyrolysis reactor and a second stage catalysis reac-tor (without any steam present). They reported that less gas yieldwas obtained for the pyrolysis of PS (6.8 wt.%) compared with thepyrolysis of PE (about 15 wt.%), when the second stage catalysttemperature was 600 �C with the presence of Y-zeolite. Lowergas yield and higher oil yield were also reported for the pyrolysisof PS than PE and PP by Encinar and González [15] in a cylindricalstainless-steel atmosphere pressure reactor at a pyrolysis temper-ature of 800 �C.
When the gasification temperature in the second stage reactorwas increased from 800 to 850 �C for the non-steam non-catalyticpyrolysis–gasification of PP, PS, HDPE and waste plastic, it appearedthat the oil yield decreased and the solid yield increased for the pyro-lysis–gasification of the three plastics (Tables 1 and 2). For example,almost no oil was collected for the non-steam, non-catalytic, pyroly-sis–gasification of PP and HDPE at the gasification temperature of
Table 1Pyrolysis–gasification of plastics; gasification temperature 800 �C, plastics weight 1.0 g, catalyst weight 0.5 g.
a The reacted water amount during each experiment.
3024 C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032
850 �C, while 24.3 and 9.7 wt.% of oil yield related to the mass of rawplastic was obtained for PP and HDPE at 800 �C, respectively. Com-pared to the pyrolysis–gasification of PP and HDPE, some oil fractionwas still collected at the gasification temperature of 850 �C for thenon-steam and non-catalytic pyrolysis–gasification of the real-world waste plastic; this might be due to some PS fractions beingpresent in the waste plastic. From Tables 1 and 2, it seems that theproduct distributions for PP and PS are much more sensitive to thegasification temperature than HDPE, when no steam and no catalystwere used. The same conclusion was also reported by Westerhoutet al. [16].
It is suggested that the carbonization reactions (endothermic)were promoted at high gasification temperature. More gaseousproducts derived from pyrolysis of the plastic were cracked intosmall molecular weight chemicals and polymerization reactionsgenerated more char. In this paper, the solid product includes res-idue from the plastic, char and coke deposited on the reactor walland catalyst. It has to be pointed out that only a little residue wasobtained for each experiment, most of the solid products were charand coke on the catalyst.
3.1.2. The introduction of steam for the non-catalytic pyrolysis–gasification of plastics
Steam was introduced into the reaction system for the pyroly-sis–gasification of plastics. The results for the steam pyrolysis–gas-ification of PP, PS, HDPE and Fost Plus waste plastic at gasificationtemperatures of 800 and 850 �C are shown in Tables 1 and 2,respectively.
The introduction of steam into the non-catalytic pyrolysis–gas-ification of PP, PS, HDPE and waste plastic appears to increase thegas yield and reduce the solid yield at both gasification tempera-tures of 800 and 850 �C (Tables 1 and 2). The oil yield was reducedwith the introduction of steam for the non-catalytic pyrolysis–gas-ification of all plastics except for PS at the gasification temperatureof 800 �C. It is suggested that thermal cracking of large molecularchemicals during the gasification reactor was influenced by theaddition of steam.
There are few reports about the influence of steam on the gasyield for the thermal decomposition of plastics. However, the influ-ence of steam addition on gas production has been reported for thethermal decomposition of other hydrocarbon materials, such asbiomass. For example, the gas yield was increased from 1.1 to1.6 Nm3 kg�1 daf biomass with increasing steam ratio, for an inves-tigation of hydrogen production from biomass gasification con-ducted at 800 �C by Turn et al. [17].
3.1.3. Hydrogen production for the non-catalytic pyrolysis–gasificationof plastics
In this section, the gas compositions and the hydrogen produc-tion from non-catalytic pyrolysis–gasification of PP, PS, HDPE andFost Plus waste plastic are presented and discussed. The gas com-positions from the pyrolysis–gasification of different plastics atgasification temperatures of 800 and 850 �C are presented in Figs.1 and 2, respectively. The H2 yields corresponding to the mass ofraw plastic at the gasification temperatures of 800 and 850 �C areshown in Tables 1 and 2, respectively.
0
10
20
30
40
50
60
70
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Gas
com
posi
tion
(Vol
.%)
Gas
com
posi
tion
(Vol
.%)
PP
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
0
10
20
30
40
50
60
70 PS
0
10
20
30
40
50
60
70
Gas
com
posi
tion
(V
ol.%
)
HDPE
Gas
com
posi
tion
(Vol
.%)
CO H
2
CO2
CH4
C2-C
4
0
10
20
30
40
50
60
70 Waste
Fig. 1. Gas compositions for the pyrolysis–gasification of plastics at a gasification temperature of 800 �C.
0
10
20
30
40
50
60
70
Gas
com
posi
tion
(V
ol.%
)
Gas
com
posi
tion
(V
ol.%
)
PP
0
10
20
30
40
50
60
70
PS
0
10
20
30
40
50
60
70
Gas
com
posi
tion
(V
ol.%
)
HDPE
Gas
com
posi
tion
(V
ol.%
)
CO H
2
CO2
CH4
C2-C
4
0
10
20
30
40
50
60
70
Waste
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Ni-Mg-Al + Steam
Sand+ Steam
SandNo Steam
Fig. 2. Gas compositions for the pyrolysis–gasification of plastics at a gasification temperature of 850 �C.
C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032 3025
From Fig. 1, 19.6 vol.% of H2, 32.7 vol.% of CH4 and 47.7 wt.% ofC2–C4 were obtained for the pyrolysis–gasification of PP withoutsteam and without Ni–Mg–Al catalyst at the gasification tempera-ture of 800 �C. Higher H2 concentration was obtained for the non-steam non-catalytic pyrolysis–gasification of PS, compared withthe PP and HDPE. However, the low H2 production (0.014 g H2/gPS) (Table 1) which was observed for the non-catalytic pyrolysis–gasification of PS, might be due to the lower overall gas yield.
From Figs. 1 and 2, it seems that the introduction of steam intothe non-catalytic pyrolysis–gasification of plastics reduced the H2
concentration at both gasification temperatures of 800 and850 �C. From Tables 1 and 2, the introduction of steam has littleinfluence on H2 production for the non-catalytic pyrolysis–gasifi-cation of plastics. In this paper, it appears that the carbonizationreactions were limited by the introduction of steam and generatedmore C1–C4 gases, CO and CO2; thus the relative concentration of
3026 C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032
H2 was reduced by the introduction of steam for the non-catalyticpyrolysis–gasification of PP and HDPE. Since a large amount of oilyield was still obtained for the non-catalytic pyrolysis–gasificationof PS at the gasification temperature of 850 �C, the influence ofsteam addition on the gas composition might be much more com-plicated for PS than for PP and HDPE (no oil produced at 850 �C).
The increase of gasification temperature from 800 to 850 �C re-sulted in a higher H2 concentration and yield for the pyrolysis–gas-ification of all the plastics. The increase of H2 concentration hasbeen reported by He et al. [18] when they investigated the produc-tion of syngas from the catalytic gasification of waste polyethylenewhen the gasification temperature was increased from 700 to900 �C.
3.2. Effect of Ni–Mg–Al catalyst on the mass balances and hydrogenproduction
3.2.1. Mass balance with Ni–Mg–Al catalystThe Ni–Mg–Al catalyst was prepared in this paper to investigate
the influence on the formation of hydrogen from the pyrolysis–gas-ification of the various plastics and the real-world Fost Plus plasticwaste. The mass balances and the product yields at the gasificationtemperatures of 800 and 850 �C are presented in Tables 1 and 2,respectively. Tables 1 and 2 show that the gas yield increased dra-matically in the presence of the Ni–Mg–Al catalyst for the pyroly-sis–gasification of the various plastics. For example, the gas yieldcorresponding to the mass of plastic increased from 76.6 to 210.9wt.% for PP, from 15.4 to 191.9 wt.% for PS, from 100.7 to 218.1wt.% for HDPE, and from 55.0 to 201.9 wt.% for the waste plastic,when the Ni–Mg–Al catalyst was introduced at the gasificationtemperature of 800 �C. It has to be pointed out that the largeamount of oil produced in the non-catalytic pyrolysis–gasificationof PS disappeared in the presence of Ni–Mg–Al catalyst. The differ-ences in gas yield for the non-catalytic pyrolysis–gasification of thedifferent plastics were reduced by the introduction of Ni–Mg–Alcatalyst. The increase of gas yield from the pyrolysis–gasificationof plastics due to the presence of catalyst has been extensively re-ported by several researchers [14,19,20].
Compared to the catalytic steam pyrolysis–gasification of PPand PS, a higher gas yield was obtained for HDPE at both gasifica-tion temperatures of 800 and 850 �C (Tables 1 and 2). The highestamount of water was consumed for the catalytic steam pyrolysis–gasification of HDPE. For example, 1.44 g water/g HDPE was con-sumed for the catalytic steam pyrolysis–gasification of HDPE atthe gasification temperature of 850 �C, compared to 1.22 g waterconsumed for 1 g of PP and 1.13 g of water consumed for 1 g ofPS. Only 1.10 g of water was reacted for the catalytic steam pyro-lysis–gasification of the Fost Plus waste plastic.
The data from Tables 1 and 2, suggest that the catalytic pyroly-sis–gasification of PP, PS and HDPE were slightly improved, and thecatalytic steam pyrolysis–gasification of waste plastic was slightlyinhibited by increasing the gasification temperature from 800 to850 �C. It is suggested that the product distribution from the cata-lytic steam pyrolysis–gasification of PP and PS are less sensitive tothe gasification temperature, compared to the experiments with-out the catalyst.
3.2.2. Hydrogen production with Ni–Mg–Al catalystThe introduction of catalyst has an important role in the
improvement of hydrogen production from the pyrolysis of hydro-carbon materials [21,22]. In this section, the influence of Ni–Mg–Alcatalyst on the H2 concentration and hydrogen production are dis-cussed for the pyrolysis–gasification of PP, PS, HDPE and the real-world, Fost Plus waste plastic.
The gas compositions for the catalytic steam pyrolysis–gasifica-tion of PP, PS, HDPE and waste plastic at gasification temperatures
of 800 and 850 �C are presented in Figs. 1 and 2, respectively. Inaddition, the hydrogen production at gasification temperatures of800 and 850 �C are shown in Tables 1 and 2, respectively.
From Figs. 1 and 2, it is shown that the H2 concentration was in-creased in the presence of the Ni–Mg–Al catalyst in the pyrolysis–gasification of PP, PS, HDPE and the waste plastic. For example, theH2 concentration was only 17.8 vol.% for the non-catalytic steampyrolysis–gasification of PP at the gasification temperature of800 �C and increased to 63.1 vol.% in the presence of the Ni–Mg–Al catalyst. In addition, the hydrocarbon gases (C1–C4) were dra-matically reduced in the presence of the Ni–Mg–Al catalyst. It issuggested that the Ni–Mg–Al catalyst greatly improved the decom-position of the hydrocarbons and produced more H2 gas. As shownin Table 1 and 2, the hydrogen production was also largely in-creased in the presence of the Ni–Mg–Al catalyst in the catalyticsteam pyrolysis–gasification of PP, PS, HDPE and the waste plastic.In addition, the increased hydrogen production in the presence ofcatalyst might be due to the promotion of the water gas reaction.As was shown in Table 1 and 2, solids production (residue productafter pyrolysis plus char product in the gasification) seemed to bereduced, especially at a gasification temperature of 850 �C. Thepromoted water gas reaction would enable more water be reactedand converted to H2.
In the presence of the Ni–Mg–Al catalyst, the CO and CO2 con-centrations were increased for the catalytic steam pyrolysis–gasifi-cation of PP, PS, HDPE and waste plastic (Figs. 1 and 2). Theincreased CO and CO2 concentration might be due to the increasedconsumed reacted water in the catalytic steam pyrolysis–gasifica-tion of the plastics. The increased amount of reacted water in thepresence of Ni–Mg–Al catalyst might result from the hydrocarbonsteam reforming reactions and also the water gas reactions. Since,hydrocarbon gas concentrations were reduced and solids produc-tion (char production) was reduced when the catalyst was intro-duced into the non-catalytic steam pyrolysis–gasificationprocess; thus CO, CO2 and H2 concentrations were increased.
Compared to the catalytic steam pyrolysis–gasification of PPand HDPE, lower H2 concentration and hydrogen production wasobtained for the catalytic steam pyrolysis–gasification of PS at bothgasification temperatures of 800 and 850 �C (Figs. 1 and 2 and Ta-bles 1 and 2). The highest hydrogen production (0.303 g H2/gHDPE) from catalytic steam pyrolysis–gasification of HDPE was ob-tained at the gasification temperature of 850 �C. Lower H2 concen-tration and hydrogen production were obtained for the catalyticsteam pyrolysis–gasification of the Fost Plus waste plastic, com-pared with the catalytic steam pyrolysis–gasification of PP andHDPE; however, slightly more H2 production was obtained com-pared with the catalytic steam pyrolysis–gasification of PS (Figs.1 and 2, and Tables 1 and 2). In addition, increasing the gasificationtemperature from 800 to 850 �C resulted in increasing productionof hydrogen for the catalytic steam pyrolysis–gasification of PP, PS,HDPE and waste plastic (Table 1 and 2).
The highest gas and hydrogen production for the catalyticsteam pyrolysis–gasification of HDPE might be due to its high ther-mal degradation rate in the pyrolysis stage [23]. In addition, thederived gaseous products for pyrolysis of HDPE were reportedmainly to be alkanes and alkenes that were suggested to be easierto be reformed in the gasification stage, compared with aromaticcompounds that are mainly products from pyrolysis of PS [24].The product distribution of pyrolysis–gasification of waste plasticseemed to be similar to the pyrolysis–gasification of HDPE andPP and might be due to the high content of polyolefin plastics(Tables 1 and 2). However, lower gas and hydrogen productionwere observed for the pyrolysis–gasification of waste plastic com-pared with catalytic pyrolysis–gasification of PP or HDPE. It is sug-gested that the contaminants contained in the waste plasticscontribute to the lower gas and hydrogen production; in addition,
C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032 3027
other small fractions of polymers such as polyvinyl chloride in thewaste plastic might generate pollutants that deactivate the catalystand produce lower gas and hydrogen yields.
3.3. Investigation of mixtures of PP, PS and HDPE plastics
Due to the limited capacity of the sample boat in the reactionsystem, 1 g of a mixture of the plastics (26.9 wt.% PP, 16.8 wt.%PS and 56.3 wt.% HDPE) resulted in an overflow of melted plastics.It seems that the mixed plastics expanded during the pyrolysisprocess. Therefore, a total amount of 0.5 g of mixed plastics wereused in this study to compare with the pyrolysis–gasification ofthe other investigated plastics (PP, PS, HDPE and the Fost Pluswaste plastic). In this section, 0.5 g of plastic, 0.5 g of Ni–Mg–Alcatalyst and 800 �C gasification temperature were applied to theinvestigation of the catalytic steam pyrolysis–gasification of theplastics using the two-stage reaction system.
Table 3Pyrolysis–gasification of different plastics; gasification temperature 800 �C, plasticsweight 0.5 g, catalyst weight 0.5 g.
Plastics PP PS HDPE Wasteplastic
Mixedplastics
Mass balance in relation to plastic + watera (wt.%)Gas/(plastic + water)
(wt.%)97.5 99.5 99.9 92.1 98.1
Oil/(plastic + water)(wt.%)
0.0 0.0 0.0 0.0 0.0
Solid/(plastic + water)(wt.%)
1.7 1.0 1.7 5.3 1.7
Mass balance (wt.%) 99.2 100.4 101.7 97.4 99.9
Mass balance in relation to PP only (wt.%)Gas/plastic (wt.%) 259.0 222.3 247.6 217.3 246.4Oil/plastic (wt.%) 0.0 0.0 0.0 0.0 0.0Solid/plastic (wt.%) 4.4 2.2 4.3 12.6 4.4Mass balance (wt.%) 263.5 224.5 251.9 229.9 250.8
Reacted water(g/g plastic)
1.7 1.2 1.5 1.4 1.5
H2 production(g H2/g plastic)
0.266 0.185 0.260 0.236 0.253
a The reacted water amount during each experiment.
PP PS HDPE0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
Gas
com
posi
tion
(Vol
.%)
Fig. 3. TGA and DTG results for the coked catalysts after pyrolysis–gasification of differ0.5 g.
The mass balance of the catalytic steam pyrolysis–gasificationof different plastics is shown in Table 3. From Table 3, no oil frac-tions were obtained for all the investigated plastics. Comparativelylow gas yield was obtained for the catalytic steam pyrolysis–gasi-fication of PS and waste plastic. The mixed plastics show a high gasyield which might be due to its high fraction of PP (26.9 wt.%) andHDPE (56.3 wt.%). In addition the amount of reacted water for themixed plastics was higher than PS and the Fost Plus waste plastic(Table 3).
As demonstrated in Tables 1 and 3, the decrease of sampleweight seems to increase the gas yield and the amount of reactedwater. It is suggested that less gaseous products were generatedduring the degradation of the sample in the first stage with thereduction of sample weight; thus comparatively more catalyticsites would be available for the gasification reactions in thesecond stage and result in higher gas yield and the amount ofreacted water. The increase of catalyst to sample ratio has beenreported to increase gas production and reacted water by Wuand Williams [9] during the catalytic steam pyrolysis–gasificationof polypropylene.
The gas compositions produced from the catalytic steam pyro-lysis–gasification of PP, PS, HDPE, Fost Plus waste plastic andmixed plastics are shown in Fig. 3. From Fig. 3, more than 60vol.% of H2 concentration was obtained for the catalytic steampyrolysis–gasification of researched plastics, except for the cata-lytic steam pyrolysis–gasification of PS. The gas compositions frommixed plastics show slight differences for the PP and HDPE plastics.The production of hydrogen (0.253 g H2/g mixed plastics) from thecatalytic steam pyrolysis–gasification of mixed plastics seems to beclose to that produced from PP and HDPE (Table 3). The hydrogenproduction from waste plastic processing is higher than from PS,but lower than from PP, HDPE and mixed plastics (Table 3).
3.4. Characterization of the coked catalysts from catalytic steampyrolysis–gasification of plastics
Temperature-programmed oxidation (TPO) analysis was carriedfor the coked Ni–Mg–Al catalysts from the pyrolysis–gasification ofthe various plastics. The results of TGA-TPO and DTG-TPO for thecoked catalysts derived from catalytic steam pyrolysis–gasificationof PP, PS, HDPE and waste plastic at the gasification temperature of
Waste Plastic Mixed Plastics
CO H
2
CO2
CH4
C2-C
4
ent plastics; gasification temperature 800 �C, plastics weight 1.0 g, catalyst weight
3028 C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032
800 and 850 �C are shown in Figs. 4 and 5, respectively, where 1 gof plastic sample was used in these experiments.
From Figs. 4 and 5, there is a maximum mass loss peak ataround 100 �C in the DGA-TPO curve for each investigated coked
200 400
-0.0004
-0.0002
0.0000
0.0002
0.0004
200 400
0.96
0.98
1.00
1.02
1.04
Polystyrene
Der
ivat
ive
wei
ght (
°C-1)
Temp
High
Wei
ght r
atio
Fig. 4. Gas compositions for the pyrolysis–gasification of PP, PS, HDPE, waste plastic and m0.5 g.
200 400
-0.0004
-0.0002
0.0000
0.0002
0.0004
0.0006
200 400
0.96
0.98
1.00
1.02
1.04
1.06
Der
ivat
ive
wei
ght (
°C-1)
Temp
Wei
ght r
atio
Fig. 5. TGA and DTG results for the coked catalysts after pyrolysis–gasification of differ0.5 g.
Ni–Mg–Al catalyst. This is likely to be due to the vaporization ofwater absorbed by the coked catalyst. Surprisingly, a peak rep-resenting an increase in mass was also obtained at around275 �C for the TPO of all the coked Ni–Mg–Al catalysts. It is
ent plastics; gasification temperature 850 �C, plastics weight 1.0 g, catalyst weight
C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032 3029
suggested that the mass increasing was due to the oxidation ofthe Ni–Mg–Al catalyst during in the TPO experiment. Since, inour previous study [22], a NiO phase was observed in the freshNi–Mg–Al catalyst, and a Ni phase was found in the coked cat-alyst determined by the X-ray diffraction analysis. It has beenreported that the nickel catalyst could be reduced by a reducingagent such as H2 during the pyrolysis/gasification experiments[25]. Therefore, it is suggested that the Ni phases in the cokedcatalysts will be oxidized and therefore gain mass during theTPO experiments.
Fig. 6. SEM results for the PP, PS, HDPE and waste pla
The mass loss after 400 �C in the TGA-TPO curve for all theinvestigated catalysts is likely to be due to coke combustion, repre-senting the amount of coke deposited on the reacted Ni–Mg–Alcatalyst. Figs. 4 and 5 indicate that the reacted Ni–Mg–Al catalystderived from catalytic steam pyrolysis–gasification of PP and wasteplastic has more deposited coke than the catalytic steam pyrolysis–gasification of PS and HDPE at both gasification temperatures of800 and 850 �C.
From Fig. 4, the main coke combustion peak occurred for theNi–Mg–Al catalyst from the experiments with PP at around
stics; plastics weight 1.0 g, catalyst weight 0.5 g.
3030 C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032
620 �C, which is consistent with out previous data [22]. However,the main coke combustion peak for the reacted Ni–Mg–Al catalystfrom the catalytic steam pyrolysis–gasification experiments withHDPE was around 550 �C, and 500 �C for the PS (Fig. 4). Similar re-sults were observed for the coked catalysts derived from the gasi-fication temperature of 850 �C (Fig. 5). It is suggested that theproperties of the coke deposited on the Ni–Mg–Al catalyst weredifferent from the catalytic steam pyrolysis–gasification of PP, PSand HDPE. The coke oxidation peak occurred at higher temperatureand could be assigned to the formation of filamentous carbonsdeposited on the catalyst [26]; this has been confirmed by SEManalysis (Fig. 6), where large amount of filamentous carbons wereobserved on the surfaces of the coked catalysts.
From the SEM analysis (Fig. 6), filamentous carbons were dif-ficult to be observed for the reacted Ni–Mg–Al catalyst derivedfrom the catalytic steam pyrolysis–gasification of PS. It mightbe suggested that the main coke combustion peak for the cokedNi–Mg–Al catalyst (from PS) could be assigned to layered car-bons [27] that have a comparatively low oxidation temperaturein TPO (Figs. 5 and 7). It has been reported that the main prod-ucts of pyrolysis of PS in the first stage were aromatics, and al-kanes and alkenes are the main products of pyrolysis of PP andPE [24]. Therefore, heavier products were expected to enter intothe second gasification stage for the catalytic steam pyrolysis–gasification of PS; additionally, more oil products were obtainedfor the non-catalytic pyrolysis–gasification of PS in this paper(Table 1 and 2). Therefore, we suggest that more layered carbons(reactive carbons) were generated from reforming heavier hydro-carbon compounds from pyrolysis of PS, compared with PP orHDPE (Figs. 7 and 8). Filamentous carbons were reported to beconverted from layered carbons [27]; thus the filamentous for-mation rate was reduced by the higher amounts of precursors(layered carbons) during the catalytic steam pyrolysis–gasifica-tion of PS, compared with catalytic pyrolysis–gasification of PPor HDPE. In addition, the higher content of layered carbons
200 400
-0.0004
-0.0002
0.0000
0.0002
0.0004
0.0006
200 400
0.96
0.98
1.00
1.02
1.04
1.06
Polystyrene
High Density PolyethyleneDer
ivat
ive
wei
ght (
°C-1)
Tempe
Hig
Wei
ght r
atio
Fig. 7. TGA and DTG results for the coked catalysts after pyrolysis–gasification of differ0.5 g.
resulted in a faster deactivation of the Ni–Mg–Al catalyst duringpyrolysis–gasification of PS and finally resulted in a lower gasand hydrogen production (Table 3).
From Figs. 4 and 5, the highest oxidation peak (at around660 �C) was obtained for the TPO experiment of the coked catalystderived from the catalytic steam pyrolysis–gasification of the FostPlus waste plastic. In addition, the diameter of the filamentous car-bons formed from waste plastic reaction appears to be larger thanthose formed from PP and HDPE catalytic steam pyrolysis–gasifica-tion (Fig. 6).
When 0.5 g of plastic was used for the catalytic steam pyroly-sis–gasification of plastics, the TPO experiment and the SEM anal-ysis were also applied to the reacted Ni–Mg–Al catalysts. The TGA-TPO and DTG-TPO results of the reacted catalysts derived from thecatalytic steam pyrolysis–gasification of PP, PS, HDPE, Fost Pluswaste plastic and mixed plastics are shown in Fig. 7. In addition,the SEM results with different magnifications for the coked cata-lysts derived from the catalytic steam pyrolysis–gasification ofPP, PS, HDPE, Fost Plus waste plastic and mixed plastics are shownin Fig. 8.
From Fig. 7, the highest amount of coke formation was observedfor the coked catalyst derived from the catalytic steam pyrolysis–gasification of the waste plastic, and the highest oxidation peakwas also obtained for the waste plastic. The DTG-TPO results forthe coked catalysts from mixed plastics present similar character-istics with those produced from the catalytic steam pyrolysis–gas-ification of PP and HDPE. The highest coke deposition on the Ni–Mg–Al catalyst for the Fost Plus waste plastic might be due tothe comparatively more serious deactivation of catalyst which re-sulted from contaminants and other toxic elements such as Cl inthe waste plastic.
The lower magnification SEM results for the coked catalystshows that there might be surface break-up of the Ni–Mg–Al cat-alyst during the catalytic steam pyrolysis–gasification of theplastics. The surface break-up mechanism for the formation of
600 800
600 800
Mixed plastics
Waste plasticsPolypropylene
rature (°C)
Mixed plasticsWaste plastics
h Density Polyethylene
Polystyrene
Polypropylene
ent plastics; gasification temperature 800 �C, plastics weight 0.5 g, catalyst weight
Fig. 8. SEM results for the PP, PS, HDPE, mixed plastics and waste plastics; plastics weight 0.5 g, catalyst weight 0.5 g.
C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032 3031
filamentous carbons on the surfaces of nickel catalysts was re-ported by Jeong and Lee [28]. From Fig. 8, the break-up surfacesof the reacted catalysts seem to be associated with the filamen-tous carbons, which were shown in the high magnifications ofthe SEM.
4. Conclusions
In this paper, PP, PS, HDPE, real-world waste plastic (FostPlus, Belgium) and mixed plastics were investigated with or
without a Ni–Mg–Al catalyst for hydrogen production from pyro-lysis–gasification in a two-stage reaction system. The main con-clusions were:
(1) Much lower gas yield (11.6 wt.% related to the mass of plas-tic) was obtained for the non-steam non-catalytic pyrolysis–gasification of PS, compared with PP (59.6 wt.%), HDPE (53.5wt.%) and waste plastic (45.5 wt.%) at the gasification tem-perature of 800 �C.
3032 C. Wu, P.T. Williams / Fuel 89 (2010) 3022–3032
(2) At the gasification temperature of 850 �C, almost no oil wascollected for the non-steam non-catalytic pyrolysis–gasifica-tion of PP and HDPE; however, 50.2 wt.% of oil for PS and12.0 wt.% for waste plastic were obtained at the same condi-tions. Increasing the gasification temperature from 800 to850 �C appears to be more sensitive for the non-catalyticpyrolysis–gasification of PP than HDPE.
(3) With the presence of the Ni–Mg–Al catalyst, the catalyticsteam pyrolysis–gasification of PP, PS, HDPE and Fost Pluswaste plastic were dramatically improved in terms of gasyield and hydrogen production. Compared to the catalyticsteam pyrolysis–gasification of PP, PS and waste plastic,higher gas yield, a higher amount of consumed water, andhigher hydrogen production were obtained for the catalyticsteam pyrolysis–gasification of HDPE at the gasificationtemperature of 850 �C. The mixed plastics (PP, PS and HDPE)showed a similar products distribution to that of HDPE, andexhibited a higher gas yield and higher amount of consumedwater compared to the PS and real-world waste plasticsamples.
(4) Filamentous carbons, suggested to be assigned to the higheroxidation temperature in the DTG-TPO analysis, wereobserved from the SEM analysis of the coked Ni–Mg–Al cat-alyst after pyrolysis–gasification of PP and HDPE. However,the filamentous carbons were difficult to find in the cokedNi–Mg–Al catalyst from the catalytic steam pyrolysis–gasifi-cation of PS at both gasification temperatures of 800 and850 �C. The type of carbons (which might be monoatomiccarbons or encapsulating carbons) deposited on the Ni–Mg–Al catalyst derived from the catalytic steam pyrolysis–gasification of PS are suggested to result in a lower produc-tion of hydrogen due to catalyst deactivation by this type ofcarbon. The highest amount of coke deposited on the surfaceof the catalyst and highest oxidation temperature for theTPO experiment were observed for the reacted Ni–Mg–Alcatalyst derived from catalytic steam pyrolysis–gasificationof the real-world waste plastic. The reacted catalyst frommixed plastics showed similar characteristics compared tothose derived from PP and HDPE.
Acknowledgements
The authors are grateful for the financial support of the Over-seas Research Student Award Scheme (UK) and the InternationalResearch Studentships Scheme (University of Leeds). The authorswould also like to thank the technical support from Mr. Ed Wood-house and the analytical support from Dr. Jude Onwudili. Grantsupport via EPSRC EP/D053110/1 is also gratefully acknowledged.
References
[1] Williams PT. Waste treatment & disposal. Chichester: John Wiley & Sons; 2005.[2] Kiran N, Ekinci E, Snape CE. Resour Conserv Recy 2000;29:273–83.[3] Pinto F, Costa P, Gulyurtlu I, Cabrita I. J Anal Appl Pyrol 1999;51:57–71.[4] Williams PT, Bagri R. Int J Energ Res 2004;28:31–44.[5] Williams PT, Williams EA. Energ Fuel 1999;13:188–96.[6] He MY, Xiao B, Hu ZQ LSM, Guo XJ, Luo SY. Int J Hydrogen Energ
2009;34:1342–8.[7] Czernik S, French RJ. Energ Fuel 2006;20:754–8.[8] Wu C, Williams PT. Appl Catal B: Environ 2009;87:152–61.[9] Wu C, Williams PT. Energ Fuel 2008;22:4125–32.
[10] Muradov NZ, Veziroglu TN. Int J Hydrogen Energ 2008;33:6804–39.[11] Lokey E. Int J Hydrogen Energ 2007;32:1673–9.[12] Williams PT, Williams EA. J Anal Appl Pyrol 1999;51:107–26.[13] Garcia L, Benedicto A, Romeo E, Salvador ML, Arauzo J, Bilbao R. Energ Fuel
2002;16:1222–30.[14] Bagri R, Williams PT. J Anal Appl Pyrol 2002;63:29–41.[15] Encinar JM, González JF. Fuel Process Technol 2008;89:678–86.[16] Westerhout RWJ, Kuipers JAM, Van Swaaij WPM. Ind Eng Chem Res
1998;37:841–7.[17] Turn S, Kinoshita C, Zhang Z, Ishimura D, Zhou J. Int J Hydrogen Energ
1998;23:641–8.[18] He M, Xiao B, Hu Z, Liu S, Guo X, Luo S. Int J Hydrogen Energ 2009;34:1342–8.[19] Uemichi Y, Nakamura J, Itoh T, Sugioka M, Garforth AA, Dwyer J. Ind Eng Chem
Res 1999;38:385–90.[20] Uemichi Y, Hattori M, Itoh T, Nakamura J, Sugioka M. Ind Eng Chem Res
1998;37:867–72.[21] Gao N, Li A, Quan C, Gao F. Int J Hydrogen Energ 2008;33:5430–8.[22] Wu C, Williams PT. Appl Catal B: Environ 2009;90:147–56.[23] Lee K, Noh N, Shin D, Seo Y. Polym Degrad Stab 2002;78:539–44.[24] Williams PT, Slaney E. Resour Conserv Recy 2007;51:754–69.[25] Clause O, Gazzano M, Trifiro F, Vaccari A, Zatorski L. Appl Catal
1991;73:217–36.[26] Wang S, Lu GQ. Ind Eng Chem Res 1999;38:2615–25.[27] Wu C, Williams PT. Appl Catal B: Environ 2010;96:198–207.[28] Jeong N, Lee J. J Catal 2008;260:217–26.
