CHINA PARTICUOLOGY Vol. 4, No.

2, 80-82, 2006

CATALYTIC CONVERSION OF MUNICIPAL WASTE PLASTIC

INTO GASOLINE-RANGE PRODUCTS OVER MESOPOROUS
MATERIALS
Jorge Norberto Beltramini
ARC Centre for Functional Nanomaterials
The University of Queensland, Brisbane, Queensland 4072, Australia
E-mail: [email protected]

Abstract In the last 20 years, it has become apparent that waste produced from plastics was becoming an envi-
ronmental problem because of their low biodegradability. Though several methods have been proposed for recycling
waste plastics, it is generally accepted that material recovery is not a long-term solution to the present problem, and that
energy or chemical recovery is a more attractive alternative, including cracking into the monomer constituents, combus-
tion to produce energy, and thermal or catalytic conversion to produce useful intermediate chemicals.
This paper is a contribution in the area of the last option for energy recovery. There have been a number of publications
reporting the use of molecular sieves and amorphous silica-alumina catalysts for the cracking of polymers into a range of
hydrocarbons. The research work reported here demonstrates the ability of mesoporous catalysts in cracking polyethyl-
ene into gasoline range products.
It was found that for mesoporous MCM-41 catalysts, its cracking activity increases with its crystallinity, displaying
higher activity with smaller pore diameters. The hydrocarbon product distribution strongly indicates a carbenium ion
cracking mechanism. The product distribution was also compared with those obtained from thermal cracking tests.
Keywords waste plastic, mesoporous materials, polyethylene cracking, MCM-41

1. Introduction ported on high surface-area carriers. Their own high acid

strength in catalyzing organic reactions in addition to their
The recycling of polymer waste is important in the con- conjugate base play a significant role in activating the re-
servation of resources and the protection of environment. actant species. When HPA's are used as heterogeneous
Thermal degradation has been studied to characterize and catalysts, specifically in gas phase reactions, the nature of
determine the stability of polymers (Garforth et al., 1997). the reactants often determines whether the reaction takes
Unfortunately, the products of thermal degradation from place on the surface or in the bulk of the catalyst.
many polymers are distributed over wide ranges of carbon This work demonstrates the ability of MCM-41-supported
numbers and their commercial value is low, requiring up- HPA's in cracking polyethylene into a range of hydrocar-
grading for effective utilization. bons of commercial importance.
On the other hand, catalytic degradation offers the po-
tential for selective recovery of useful chemical fractions by 2. Experimental
influencing the product distribution. To date, the catalytic
degradation of polymeric materials has centred around the 2.1 Catalyst preparation
active components in commercial fluidized catalytic crack- The MCM-41 samples used were synthesized using the
ing, namely zeolites Y and ZSM-5. One problem with hydrothermal crystallization technique (Izumi et al., 1989),
polymer degradation over these microporous zeolite cata-
in which cetylmethylammonium bromide was dissolved in
lysts is the limited access to the active catalytic sites via
distilled water and an aqueous solution of tetramethylam-
pore opening of diameter of 7.4 A in zeolite Y and 5.1 A in
monium hydroxide was added while stirring, followed by a
ZSM-5.
silica solution. Then the hydrogel was crystallized at 160°C
An important new class of molecular sieves, called
for 120 hours in a pressure vessel. The crystalline product
M41S, has been reported in the literature with interesting
properties such as large pore volume and size and high was filtered, washed, dried at 100°C and finally calcined at
thermal stability. Clearly, there is an excellent potential for 500°C for 18 hours. In impregnation, the HPA (Heter-
these materials in commercial catalysis, especially where opolyacid acid) was dissolved in distilled water and cal-
larger pore size than those offered by the conventional cined MCM-41 was then added to it. The mixture was
zeolite is required. It has been shown that the MCM-41 stirred for 18 hours and then left at 100°C in an oven until
materials were unlikely to possess strong acid sites as in the water evaporated. This procedure gave a loading of
zeolites (Corma et al., 1995). The impregnation of heter- 16.6% by weight of HPA on the supported catalyst.
opolyacids (HPA's) on MCM-41 materials as a means of
2.2 Catalyst characterization
increasing acidity was lately reported (Diaz et al. 1999).
HPA's are well known superacids especially when sup- X-ray powder diffraction measurements were carried out
 Beltramini: Catalytic Conversion of Municipal Waste Plastic over Mesoporous Materials 81

using a Siemens D5005 powder X-ray diffraction instru- MCM-41. The product composition of the cracking reaction
ment. Diffraction patterns were recorded for 2-θ angle be- is a direct indication of the catalyst activity and selectivity.
tween 0.4° and 30°. Adsorption measurements were car- The composition of the liquid shifts toward lower carbon
ried out on a Hidden Analytical IGA gravimetric system. number as the catalyst activity increases. The gas chro-
Small pulses of benzene were introduced periodically via a matograms demonstrated that the liquid product obtained
gas admittance valve which was electronically controlled to from the cracking of polyethylene over HPW/MCM-41 is
obtain adsorption steps of 2% by weight. Sample weight much lighter than the corresponding non-catalytic test car-
was recorded on a highly accurate micro-balance, and ried out at the same temperature, as shown in Fig. 1 for the
pressure of adsorbate was measured using a set of pres- quantitative yields of products in the C1-C25 fraction as
sure transducers. Data processing was handled by the use function of the carbon number.
of a computer. 40

Thermal cracking
2.3 Catalytic tests 35
HPW/MCM-41
MCM-41
Polyethylene cracking was carried out under atmos- 30

pheric pressure in a stainless steel batch reactor which

Yield / wt%
25

was connected to a liquid condenser and a gas collection

20
bag. Gaseous product was analysed using a GC with FID
and a KCl/Al2O3 capillary column. Liquid product was ana- 15

lysed also using a GC with FID and SE30 capillary column. 10

The apparatus containing the polyethylene and the catalyst
5
was flushed with nitrogen for at least 10 minutes. The re-
actor was then placed in a furnace at the reaction tem- 0
0 5 10 15 20 25 30
perature. The volume of the liquid collected was measured Carbon Number of Components
periodically using a graduated cylinder. The reaction was Fig. 1 Yield (wt%) of cracking product of high density polyethylene
considered complete when no further liquid was collected over different catalysts at 450°C.
over a period of an hour. The residue which remained with
the catalyst and the liquid product were weighed after the As can be seen in the figure, the lightest products were
completion of the experiment. The weight of gas was cal- obtained over HPW/MCM-41, followed by pure silica MCM-
culated by difference. 41. Both catalysts give substantially higher yields in the
C4-C8 fraction than in the case of non catalytic thermal
3. Results and Discussion cracking. The yields of C1-C8 product fraction of all crack-
ing tests are reported in detail in Table 2.
MCM-41 with all silica composition was synthesized with
standard unit cell parameters of 60 A. Characterization Table 2 Reaction conditions and composition of cracking products
results from X-ray diffraction and benzene adsorption, Reaction Product / wt% C1-C8
showed that the material had an all silica composition with Catalyst tempera-
Gas Liquid Residue / wt%
an aluminium impurity level lower than 10 ppm and with a ture / °C
typical hexagonal unit cell symmetry. Upon impregnation MCM-41 450 8.5 82.4 9.1 56.6
with tungstophosphoric acid (HPW) the unit cell of the HPW/MCM-41 450 10.4 86.2 3.4 87.3
MCM-41 material shrank by 12 A. This reduction appeared None 450 6.3 81.6 12.1 29.2
to have occurred primarily in the wall structure, as the drop
in the benzene adsorption capacity did not fully account for 4
the extent of the unit cell shrinkage (Blasco et al., 1998).
3.5
These properties are shown in Table 1.
3
Liquid collected / mL

Table 1 Properties of MCM-41 sample synthesized

2.5
Crystalliza- d100 Vp
a0 / A P/Po dp / A δ/A 2
tion time / h /A / cm3⋅g-1
110 52.8 61.7 0.72 0.36 50.9 10.9 1.5

d100: diffraction unit cell index; a0: unit cell; Vp: pore volume; P/Po: BET 1
pressure; dp: pore diameter; δ: wall thickness. HPW/MCM-41
0.5 MCM-41

3.1 Polyethylene cracking 0

0 20 40 60 80 100 120 140
The cracking of high density polyethylene (HDPE) was Time / min
studied by both pure thermal cracking, and catalytic Fig. 2 Rate of liquid collection from polyethylene cracking reaction
cracking on pure MCM-41 and on HPW impregnated over HPW/MCM-41 and silica MCM-41.
82 CHINA PARTICUOLOGY Vol. 4, No. 2, 2006

The first indication of the apparent cracking rate of a The product distribution is also indicative of the reaction
catalyst is the rate of liquid collection. The catalytic crack- mechanism. It has been claimed that the cracking of poly-
ing rate is always higher than the thermal cracking rate ethylene over all silica MCM-41 occurs via free radical
without any catalyst at the same temperature. In both, mechanism and that the pore structure acts as a reaction
catalytic and non-catalytic thermal tests, the rate of liquid vessel stabilizing the radicals in comparison to thermal
collection increases with increasing the time of run as cracking in the absence of a catalyst (Sakata et al., 1996).
shown in Fig. 2. Higher rate of liquid yields over However, no product distribution has been reported in
HPW/MCM-41 rather than on MCM-41 were also con- support of this proposed mechanism. It is evident from
firmed from this experiments. These results may be attrib- Table 3, however, that high yields of isobutene and isobu-
uted to the higher surface area and smaller pore diameter tane and low levels of methane and ethane always ac-
of HPW/MCM-41. company high activity of the mesoporous catalysts.

Table 3 Composition of gaseous product

Gaseous product distribution / wt%

Catalyst (iC4P / nC4P) (i C = / n C = )
4 4
C1 C2 C2= C3 C3= C4 C 4=
MCM-41 0.16 0.50 0.23 0.78 2.34 0.80 3.62 0.95 0.51
HPW/MCM-41 0.04 0.11 0.08 0.30 2.15 0.72 3.60 5.31 0.49
None 0.44 1.29 0.44 1.75 1.94 1.01 1.26 0.00 0.09

The cracked products contain high levels of methane, pared with pure MCM-41 test runs. It is proposed that the
ethane, ethene and olefins relative to the isobutene and formation and stabilization of carbenium ion in the pores of
isobutane levels. This scheme is in accord with the product the MCM-41 might be due to the adsorption between the
distribution of the non-catalytic run reported in Table 3. polyethylinic fragments with the surface of the channels
However, the high levels of isobutene and isobutane in the where the HPW is adsorbed.
products of the catalytic runs indicate that the reaction is New mesoporous materials, in particular enhanced-
dominated by a carbenium ion mechanism. acidity MCM-41, are confirmed to be of potential interest in
The product analysis shows that there are concentra- the cracking of heavier feedstocks. The results of the pre-
tions of long chain alkanes as well alkenes, and these can sent study confirm this suggestion since these materials
be activated via hydrogen transfer involving isobutyl car- are active in the degradation of waste plastics.
benium ion intermediate.
The findings from the polyethylene cracking reaction are References
consistent with the product distribution for the MCM-41
Blasco, T, Corma, A, Martinez, A. & Escolano, P. (1998). Sup-
catalysts as reported in Tables 2 and 3. As can be seen,
ported heteropolyacid (HPW) catalysts for the continuous alky-
the ratios of isobutene in the butenes and of isobutane in lation of isobutane with 2-butene: The benefit of using MCM-41
the butanes increase with increasing catalytic activity. In with larger pore diameters. J. Catal., 177(2), 306-313.
the case of non-catalytic cracking test, these ratios are Corma, A., Iglesia, M. & Sanchez, F. (1995). Large pore bifunc-
0.08 and 0 respectively, whereas the same ratios over the tional titanium-aluminosilicates: The inorganic non-enzymatic ver-
MCM-41 catalyzed cracking tests are as high as 0.51 and sion of the epoxidase conversion of linalool to cyclic ethers. J.
5.3 respectively. These results strongly support the pro- Chem. Soc., Chem. Commun., 1995, 1635-1636.
Dias, J. A., Osegovic, J. P. & Drago, R. S. (1999). The solid acidity
posed carbenium ion mechanism.
of 12-tungstophosphoric acid. J. Catal., 183(1), 83-90.
Garforth, A., Fiddy, S., Lin, Y. -H., Ghanbari-Siakhali, A., Sharratt,
4. Conclusion P. N. & Dwyer, J. (1997). Catalytic degradation of high density
polyethylene: An evaluation of mesoporous and microporous
It has been shown that addition of HPW to all silica catalysts using thermal analysis. Termochimica Acta, 294(1),
MCM-41 catalyzes the cracking reaction of hydrocarbons. 65-69.
The product distribution of cracking reactions strongly in- Izumi, Y., Natsume, N., Takamine, H., Tamaoki, I. & Urabe, K.
dicates a carbenium ion mechanism, associated with acid (1989). Effect on synthesis of mesoporous materials. Bull.
sites created by the HPW lining the channels of MCM-41. Chem. Soc. Japan, 62, 2159-2164.
Sakata, Y, Uddin, M. A., Koizumi, K. & Murata, K. (1996). Catalytic
The lightest product was obtained over the HPW im- degradation of polypropylene into liquid hydrocarbons using sil-
pregnated MCM-41 followed by those obtained over pure ica-alumina catalyst. Chem. Lett., 25(3), 245-246.
MCM-41. The levels of isobutane and isobutene are very
high relative to the levels of methane and ethane as com Manuscript received February 6, 2006 and accepted March 8, 2006.