Journal of the Air & Waste Management Association

ISSN: 1096-2247 (Print) 2162-2906 (Online) Journal homepage: [Link]

Systematic Evaluation and Uncertainty Analysis of

the Refuse-Derived Fuel Process in Taiwan

Ying-Hsi Chang , Ni-Bin Chang & W. C. Chen

To cite this article: Ying-Hsi Chang , Ni-Bin Chang & W. C. Chen (1998) Systematic Evaluation
and Uncertainty Analysis of the Refuse-Derived Fuel Process in Taiwan, Journal of the Air & Waste
Management Association, 48:6, 537-544, DOI: 10.1080/10473289.1998.10463702

To link to this article: [Link]

Published online: 27 Dec 2011.

Submit your article to this journal

Article views: 205

View related articles

Citing articles: 9 View citing articles

Full Terms & Conditions of access and use can be found at

[Link]
TECHNICAL PAPER ISSN 1047-3289 [Link],
Air & WasteChang, and48:537-544
Manage. Assoc. Chen
Copyright 1998 Air & Waste Management Association

Systematic Evaluation and Uncertainty Analysis of the

Refuse-Derived Fuel Process in Taiwan
Ying-Hsi Chang, Ni-Bin Chang, and W. C. Chen
Department of Environmental Engineering, National Cheng-Kung University, Tainan, Taiwan, R.O.C.

ABSTRACT INTRODUCTION
In the last few years, Taiwan has set a bold agenda in Continuous economic development and population growth
solid waste recycling and incineration programs. Not have inevitably resulted in a rapid increase of municipal
only were the recycling activities and incineration solid waste (MSW) in the past decades in many developing
projects promoted by government agencies, but the countries. Situations in different countries, however, might
related laws and regulations were continuously pro- generate different planning and policies for waste manage-
mulgated by the Legislative Yen. The solid waste pre- ment, but as a rule, the foundation for planning and de-
sorting process that is to be considered prior to the sign of any waste management system is governed by the
existing incineration facilities has received wide atten- amount and characteristics of the waste generated. Accord-
tion. This paper illustrates a thorough evaluation for ing to an investigation conducted by the Environmental
the first refuse-derived fuel pilot process from both Protection Administration in Taiwan, the generation rate
quantitative and qualitative aspects. The process is to of MSW was increasing from 18,468 tons/day in 1990 to
be installed and integrated with a large-scale munici- 23,268 tons/day in 1994.1 More and more paper and plas-
pal incinerator. This pilot process, developed by an tics exist in the waste streams. As the sources and prices of
engineering firm in Tainan County, consists of stan- raw material become critical, recyclables, which constitute
dard unit operations of shredding, magnetic sepa- more than half of the physical composition in the waste
ration, trommel screening, and air classification. A streams in Taiwan, would be a valuable resource in the sec-
series of sampling and analyses were initialized in ondary material market. On the other hand, possible sepa-
order to characterize its potentials in the solid waste ration of non-combustibles from the waste streams would
management system. The probabilistic modeling for be beneficial for the subsequent incineration or pyrolysis
various types of waste properties derived in this process. But the heterogeneity, uncertainty, and variability
analysis may provide a basic understanding of sys- of waste composition make such applications intractable.
tem reliability. Several alternative processes for the treatment of MSWs
are under development in Taiwan, of which probably the pri-
mary goal is the recovery of material. In the literature, many
studies have been focused on the evaluation of material recov-
IMPLICATIONS
ery systems. Experiments were independently conducted for
This analysis presents a thorough performance testing and
reliability analysis for the refuse-derived fuel (RDF) pro- the evaluation of various types of unit operations, such as the
cess that was built as the first pilot plant in Taiwan. The shredder,2-5 magnetic separator,6 air classifier,6-11 trommel,12-14
mass balance analysis indicates that such a presorting and the vibrating system.15 General design principles and is-
process for subsequent waste-to-energy facilities may sues relating to recovering ferrous metal, aluminum, plastics,
achieve a 50% production rate based on the waste inflow glass, and paper were discussed.16-24 However, the integration
of 30 tons/hr in several test runs. With the resultant higher
of appropriate equipment to form a workable system is a func-
heat value (30 tons/hr up to 3716 kcal/kg on average) and
lower amount of non-combustibles existing in the final RDF tion of the characteristics of waste inflow and the specifica-
products, the systematic implications to energy recovery, tions of the recovered products required in the secondary ma-
combustion efficiency, pollution control, and ash reuse terial market. The application of reliability and economic analy-
potential in an integrated waste-to-energy facility are phe- ses to the selection of a refuse-shredding system has been con-
nomenal. Reliability analysis by the identification of prob-
ducted by Bowen and Huang.25
ability distribution to each related factor extends the in-
sight in the feasibility study. Therefore, a comparative evaluation of several avail-
able options, such as building material recovery facilities
or installing solid waste presorting processes prior to the

Volume 48 June 1998 Journal of the Air & Waste Management Association 537
Chang, Chang, and Chen

Figure 1. Flowchart of the solid waste presorting process.

existing incinerators, becomes a necessary part of solid waste and ceramics, to reduce the content of dense material in
management planning. Of the many criteria that need to the residual MSW streams. Residual materials, passing
be considered in the presorting process, only technical and through an air classifier, are sent into the trommel screen
economic viabilities have received much attention. In an for advanced separation. The dimensions of the openings
attempt to start the evaluation of both the technical po- on the surface of the trommel screen can be varied to fine-
tential and economic feasibility of an MSW presorting plant, tune the processing function and ensure maximum com-
the first pilot plant, with a capacity of 30 tons/hr, was de- bustibles recovery. The trommel is designed as two concen-
signed and built by a local company in Tainan County in tric shells. The outer shell, 2.33 m in diameter and 4.3 m in
Taiwan. This pilot presorting process consists of several stan- length, has many circular holes on the surface designed to
dard units, such as shredding, magnetic separation, trommel remove shredded materials smaller than 25 mm. The inner
screening, and air classification. Several samplings and shell, 1.9 meters in diameter and 4.56 m in length, sepa-
analyses were initialized by a research group in National rates partial waste streams between 25 and 100 mm, re-
Cheng-Kung University. The principles of sound technical garded as the trommel middle flow. The rest of the waste
evaluation, particularly with regard to the inevitable un- streams are trommel overflow. Alternatively, three waste
certainty in system performance and their application to streams can be further classified in which two of them are
an available treatment option, such as incineration or py- trommel undersize. In other words, particle size is controlled
rolysis, are emphasized in this paper. The analytical outcome by the design of the openings on the surface of the trommel;
may improve the decisionmaking regarding the application of material with a particle size less than 25 mm and a particle
the MSW presorting process to waste-to-energy plants. size between 25 mm and 100 mm are separately arranged
by two different sets of openings with a concentric shell
FACILITY DESCRIPTION configuration. The overflow, passing through this trommel
As illustrated in Figure 1, the designed process consists of screen, would present the least dense portion in the MSW,
three major subsystems: shredding, air classification, and with sizes greater than 100 mm, and can be exactly identi-
screening. The facility may process 30 tons/hr on one line fied as the fluff-RDF (refuse-derived fuel) that is the end
at maximum capacity. The MSW is delivered to the facility product of this sorting process. However, outputs with par-
by packer trucks. The bag-ripping unit, in charge of open- ticle sizes larger than 100 mm and particle sizes between 25
ing the plastic bags, initializes the sorting process. Ferrous mm and 100 mm can be used as alternative fuels in the
metal is then extracted from the MSW stream using mag- waste-to-energy facilities. Every unit is fully enclosed for
nets. Recovered ferrous metal is conveyed to a ferrous stor- noise and dust control. However, odor control is not con-
age bin from which it will be recycled. The MSW is then sidered at the current stage. The price of this pilot plant
shipped into a vertical hammermill shredder, followed by was approximately $3,200,000 in 1994. Table 1 lists the
an air classifier, using a belt-type conveyor. Non-ferrous energy requirement of each unit in this presorting process.
materials, such as aluminum cans and other combustibles,
as well as the MSW, are crushed by the vertical hammermill TESTING RESULTS
shredder. The air classifier, blowing with a regular air stream A considerable amount of MSW in Tainan County was
of 200 m3/min from the vertical hammermill shredder, fur- collected, and the assessment of the presorting process
ther isolates and separates the inert materials, such as glass ran from July to December 1995. A mass balance diagram

538 Journal of the Air & Waste Management Association Volume 48 June 1998
 Chang, Chang, and Chen

Table 1. Energy requirement of each unit in the solid waste presorting process.

Equipment Capacity Energy Requirements (hp)

Bag ripper 30 (tons/hr) 5

Magnetic separator 30 (tons/hr) 3
Shredder 30 (tons/hr) 500
Air classifier 200 m3-air/min 20
Trommel 30 (tons/hr) 7.5
Cyclone 400 m3-air/min 40
Conveyors 30 (tons/hr) 1~10

showing the mass flux of waste stream throughout the

presorting process is provided in Figure 2. This diagram is
based on a processing of 21 tons/hr of MSW over a spe-
cific testing period. For each 100 tons of MSW processed,
approximately 50 tons of RDF are produced (i.e., the sum- Figure 2. Mass balance diagram of the solid waste presorting
process.
mation of trommel overflow and middle flow), and approxi-
mately 2.5 tons of ferrous metal are recovered. Thirteen operation of magnetic separation. The air classifier effec-
samples were collected at locations 1 and 6 in Figure 1, tively differentiates the heavy materials from the incom-
while 15 samples were collected at locations 3–5. ing waste streams. However, parts of flattened aluminum
The heat value, physical composition, proximate analy- and iron cans were found in the stream of dense materials.
sis, and ultimate analyses were also established in our labo- This phenomenon reveals that bag ripping might not be
ratories, as shown in Table 2. It unambiguously shows that an effective method to provide a full exposure environment
the plastics content dramatically increased from 26.33% in for subsequent magnetic separation, partially due to the
the MSW to 67.81% in the trommel overflow at dry basis. habit of using different sizes of plastic bags for various types
On the other hand, food waste, metal, glass, and ceramics of MSW in the waste storage process in Taiwan. For instance,
were reduced almost to zero in the trommel overflow. The many families store food waste in a smaller plastic bag to
water content decreased from 50.65% in the MSW to prevent odor and insect invasion and then place the smaller
40.28% in the trommel overflow, due to the evaporation plastic bag inside a larger plastic bag. In addition, the effi-
effect during the air classification process. The combustible ciency of the shredding machine is low, and the duct at the
content increased from 37.15% in the MSW to 49.76% in exit of shredder is too long to fluently transport the shred-
the trommel overflow. Similarly, the combustible content ded materials into the air classifier, so a choking phenom-
increased from 20.11% in the MSW to 29.24% in the enon occurs frequently. To improve system efficiency, the
trommel overflow. In addition, the chlorine and sulfur con- use of a better bag-ripping process and shredding machine
tents would be the major elements for the examination of are essential. A shorter duct, connecting the air classifier
the impact on air pollution control in the subsequent in- and shredder, are required as well. The ferrous metals are
cineration process. The sulfur content decreased from 0.8% separated by a magnetic separator before shredding, but
in the MSW to 0.05% in the trommel overflow, but the the non-ferrous metals should be separated manually or by
chlorine content slightly increased from 0.18% in the MSW eddy current before shredding. It is suggested that one more
to 0.23% in the trommel overflow. This is probably due to unit of manual sorting for isolating those non-ferrous metal
the increase of plastics in the trommel overflow. However, materials is required between magnetic separation and the
the higher heat value was increased from 2,277 kcal/kg in shredding machine, and a recirculated conveyor, prepared
the MSW to 3,715 kcal/kg in the trommel overflow. A higher for those incompletely opened plastic bags, would be help-
heat value increase of almost 66% in the RDF stream, based on ful to improve the recycling efficiency.
such a presorting process, would result in a higher degree of
energy recovery if a waste-to-energy facility is installed. Quantitative Evaluation
A study was performed to examine the variability of this
PERFORMANCE EVALUATION RDF production process. The probability distribution of the
Qualitative Evaluation product characteristics was independently identified, us-
In general, the testing results imply that the potentials of ing the goodness-of-fit statistical test. The Komogorov-
RDF production are remarkable. One-tenth of non-ferrous Smirnov test was eventually selected, and the lognormal
metal was found in the ferrous metal stream after the distribution was found to be the most suitable for the

Volume 48 June 1998 Journal of the Air & Waste Management Association 539
Chang, Chang, and Chen

Table 2. The mean and variance of the samples corresponding to each property at different sampling location.

MSW Heavy Material < 25mm 25~100mm > 100mm

Mean Variance Mean Variance Mean Variance Mean Variance Mean Variance

Physical composition
(on dry basis, %)
Paper 28.62% 8.57% 2.50% 1.73% 0.10% 0.24% 8.08% 3.29% 5.70% 3.00%
Plastics 26.33% 7.03% 9.00% 7.19% 1.00% 1.07% 29.15% 7.19% 57.81% 9.49%
Garden trimmings 4.05% 2.13% 5.40% 2.97% 0.40% 0.41% 4.60% 3.96% 4.21% 2.37%
Textiles 9.03% 6.90% 9.30% 7.96% 0.00% 0.00% 7.43% 3.38% 18.23% 6.71%
Food waste 14.04% 10.01% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
Leather/rubber 0.58% 1.39% 0.50% 0.76% 0.10% 0.16% 1.13% 1.54% 1.48% 2.73%
Metal 6.99% 4.56% 0.30% 0.81% 0.00% 0.00% 1.09% 1.44% 0.03% 0.11%
Glass 7.26% 8.83% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
Ceramics & china 0.47% 0.88% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
<5mm 1.59% 1.96% 35.00% 22.73% 54.80% 10.67% 16.15% 5.52% 8.89% 5.09%
>5mm 1.04% 1.97% 37.80% 11.61% 43.80% 11.99% 32.36% 15.96% 3.65% 3.59%
Heat Value
HHV(kcal/kg) 2277.8 477.2 1643.7 455.03 1376.8 310.6 2554.5 342.0 3715.9 551.6
LHV(kcal/kg) 1816.3 489.2 1183.2 484.28 936.7 331.7 2095.7 342.2 3296.0 528.4
Chemical composition
(on wet basis, %)
C 20.11% 5.00% 16.89% 3.47% 20.55% 3.51% 24.45% 4.34% 29.24% 4.58%
H 2.92% 0.29% 2.17% 0.47% 2.30% 0.44% 3.21% 1.09% 3.30% 1.54%
N 0.55% 0.41% 0.76% 0.48% 1.05% 0.51% 1.09% 0.55% 1.04% 0.85%
Cl 0.18% 0.14% 0.08% 0.04% 0.11% 0.08% 0.16% 0.15% 0.23% 0.22%
S 0.80% 2.69% 0.05% 0.05% 0.06% 0.05% 0.10% 0.11% 0.05% 0.05%
O 12.58% 3.19% 11.27% 4.46% 11.27% 3.82% 11.69% 3.53% 15.90% 9.54%
Proximate analysis
(on wet basis, %)
Moisture 50.65% 6.71% 59.20% 4.90% 52.65% 5.43% 47.55% 4.87% 40.28% 7.61%
Ash 12.21% 5.90% 11.02% 4.18% 12.00% 2.99% 11.75% 5.00% 9.96% 4.78%
Combustibles 37.15% 10.96% 29.78% 5.57% 35.35% 3.60% 40.70% 5.35% 49.76% 9.96%

illustration of product characteristics. These characteristics the trommel overflow suitable for subsequent waste-
include the heat value, physical property, chemical element, to-energy process consists of 40% moisture, 50%
and proximate content. Overall, Figures 3–6 present a se- combustibles, and 10 % ash.
ries of histograms to illustrate the paper and plastics con- • According to the goodness-of-fit statistical test, all
tents, heat values, proximate analysis, and ultimate analy- of the physical and chemical properties can be de-
sis. These displays show that the properties vary for each scribed by lognormal distributions with various
group of samples. The first and second numbers in paren- mean and variance values.
theses beside the LN symbols above the histograms are the • Since both waste streams from trommel middle flow
mean and variance values, respectively, for those sampling and overflow can be used as RDF, an integrated pre-
locations, as indicated above each diagram. diction equation can be established. Therefore, the
Overall, mass balance, physical composition, proxi- prediction capability of higher heat values is avail-
mate analysis, chemical property, heat value, and proba- able for subsequent waste-to-energy process.
bilistic modeling were presented, and several discoveries
were made: Compatibility Analysis
• According to the mass balance analysis, approxi- Thermal treatment using incineration technology has
mately 46 tons of RDF and 2.2 tons of ferrous metal been proven as an attractive method of waste disposal
are recovered from each 100 tons of municipal waste. due to the primary advantages of hygienic control, vol-
• Kw-hr is required for running this presorting pro- ume reduction, and energy recovery. Although Taiwan has
cess every day. set a bold agenda for solid waste incineration programs
• The trommel overflow presents 58% of plastics and over 22 large-scale municipal incinerators have ei-
content and 3,716 kcal/kg of higher heat value. In ther been built, are under construction, or are in the plan-
addition, proximate analysis shows that the RDF in ning stages in the last few years, the continuing subsidy

540 Journal of the Air & Waste Management Association Volume 48 June 1998
 Chang, Chang, and Chen

Figure 3. The histogram and probability distribution of plastics and paper present in stages 1-6 of the solid waste presorting process. The mean
and variance, respectively, are shown above each plot.

Figure 4. The histogram and probability distribution of HHV and LHV present in stages 1-6 of the solid waste presorting process. The mean and
variance, respectively, are shown above each plot.

Volume 48 June 1998 Journal of the Air & Waste Management Association 541
Chang, Chang, and Chen

to local government agencies for promoting waste recy- the integrated value of solid waste presorting to the
cling activities has never been tempered. A recent trend incineration, pyrolysis, or a liquefaction process would
shows that material recycling and waste-to-energy facili- present a new perspective in solid waste management.
ties may be integrated to respond to continuous landfill In the last few years, a number of credible surveys26-33
closings and to improve the quality of the incineration have shown that recycling and waste-to-energy appear to
process. At least one proposal for planning an additional work well together. Compatibility even exists for several
MSW presorting process for the large-scale municipal in- more reasons related to economic, environmental, politi-
cinerator located in Tai-Chung has been discussed in 1996. cal, and social aspects. In this analysis, RDF generated from
Such an integrated system configuration, with the MSW locations 5 and 6 in Figure 1 were found suitable for an
presorting and incineration units, has also been proposed incineration process. Besides, RDF products larger than 100
in the planning stages of several modular incinerators mm from the trommel overflow, of which plastics are the
located in several suburban areas. In addition, a waste liq- major content, are particularly useful for waste pyrolysis or
uefaction process, which is well-designed for handling the liquefaction process. However, the impacts of solid waste
scrap tires, plastic and rubber waste, and auto shredder presorting on waste-to-energy facilities remain uncertain
residues, was built in Tainan in 1996. Recognition of in different countries due to the inherent complexity of

Figure 5. The histogram and probability distribution of proximate analysis present in stages 1-6 of the solid waste presorting process. The mean

542 Journal of the Air & Waste Management Association Volume 48 June 1998
 Chang, Chang, and Chen

Figure 6. The histogram and probability distribution of ultimate analysis present in stages 1-6 of the solid waste presorting process. The mean and
variance, respectively, are shown above each plot.

Volume 48 June 1998 Journal of the Air & Waste Management Association 543
Chang, Chang, and Chen

4. Savage, G.M.; Trezek, G. J. Significance of Size Reduction in Solid Waste

solid waste composition. Further economic and technical Management; U.S. Environmental Protection Agency: Washington, DC,
feasibility studies are to be pursued in the future. 1980, Vol. I; EPA-600/2-80-115.
5. Savage, G.M.; Tuck, J. K.; Gandy, P. A.; Trezek, G. J. Significance of Size
Reduction in Solid Waste Management; U.S. Environmental Protection
CONCLUSION Agency, 1983, Vol. II; EPA-600/2-83-006.
6. Simister, B.W.; Bendersky, D. Processing Equipment for Resource Recov-
This analysis investigates the performance of Taiwan’s ery Systems: Vol. II - Magnetic Separators, Air Classifiers and Ambient Air
Emissions Tests; U.S. Environmental Protection Agency, 1980; EPA-
first solid waste presorting process, prepared to be in- 600/2-80-007b.
tegrated with subsequent waste-to-energy facilities in 7. Biddulph, M.W.; Connor, M.A. Resources, Conservation and Recycling
1989, 2, 275-286.
the future. According to the characteristics of local 8. Murray, D.L. “Air classifier performance and operating principles,”
in Proceedings of 1978 Waste Processing Conference, Chicago, IL, 1978.
solid waste, this presorting process, consisting of sev- 9. Nollet, A.R.; Hasselriis, F. “Test of rotary drum air-classifier at Albany,
eral essential unit operations, such as bag ripping, New York, solid waste energy recovery system,” in Proceedings of 1986
National Waste Processing Conference; ASME: New York, 1986.
magnetic separation, shredding, air classification, and 10. Savage, G.M.; Diaz, L.F.; Trezek, G.J. “Performance characterization
screening, was designed to isolate the non-combus- of air classifiers in resource recovery processing,” in Proceedings of
1980 Waste Processing Conference; ASME: New York, 1980.
tibles and food waste from the waste streams and gen- 11. Senden, M.M.G. “Performance of zig-zag air classifiers at low particle
concentrations: a study of the effect of stage geometry variations,” in
erate the qualified RDF for subsequent treatment and Proceedings of 1980 Waste Processing Conference; ASME: New York, 1980;
disposal. As a whole, this paper addresses a very im- pp 351-359.
12. Sullivan, J.W.; Hill, R.M.; Sullivan, J.F. “The place of the trommel in
portant topic in solid waste management in a devel- resource recovery,” in Proceedings of 1992 National Waste Processing
Conference; ASME: New York, 1992.
oping area or nation. Information regarding the qual- 13. Alter, H.; Gavis, J.; Renard, M.L. “Design models of trommels for re-
ity and quantity of various recovered waste streams source recovery processing,” in Proceedings of 1980 Waste Processing
Conference; ASME: New York, 1980.
will be valuable for the planning and design of future 14. Wheeler, P.A.; Barton, J.R.; New, R. Resources, Conservation and Recy-
recycling centers or incineration plants in Taiwan. cling 1989, 2, 261-273.
15. Musschoot, A.; Sherman, R.; Guptail, W. “Vibrating process equip-
From the mass balance diagram, we should be able to ment for resource recovery,” in Proceedings of 1992 National Waste
Processing Conference; ASME: New York, 1992.
determine how much material (by percentage) of the 16. Alter, H. Material Recovery from Municipal Waste: Unit Operation in Prac-
waste stream is recycled and what portion will be dis- tice; Marcel Dekker: New York, 1983; p 181.
17. CalRecovery and PEER Consultants. Material Recovery Facility Design
posed of by incineration or landfill. In addition, proba- Manual; CRC Press: Florida, 1993.
18. Design & Management for Resource Recovery: Vol. I - Energy from Waste;
bilistic modeling derived in this performance testing may Frankiewicz, T. C., Ed.; Ann Arbor Science: Michigan, 1980.
provide the basis for evaluating engineering reliability 19. Hecht, N.L. Design Principles in Resource Recovery Engineering;
Butterworths: Boston, 1983.
on solid waste presorting and waste-to-energy facilities. 20. Klee, A.J.; Vesilind, P.A. Design & Management for Resource Recovery: Vol. III-
However, in response to situations with higher Quantitative Decision Making; Ann Arbor Science: Michigan, 1980.
21. Savage, G.M.; Diaz, L.F. “Processing of solid waste for material recov-
moisture content and complex characteristics in the ery,” in Proceedings of 1990 National Waste Processing Conference, Cali-
fornia, 1990.
waste stream in Taiwan, further modification of this 22. Swartzbaugh, J.T.; Diaz, L.F.; Duvall, D.S.; Savage, G.M. Recycling Equip-
presorting process is required. Since the prices of ment and Technology for Municipal Solid Waste; Noyes Data Corpora-
tion: New Jersey, 1986.
recyclables in the secondary material market and the 23. Savage, G.M.; Glaub, J.C.; Diaz, L.F. Unit Operations Models for Solid
Waste Processing; Noyes Publication: Park Ridge, NJ, 1986.
composition of the waste stream are always unstable, 24. Vesilind, R.A.; Rimer, A.E. Unit Operations in Resource Recovery
the designer must fully grasp the dynamic characteris- Engineering; Prentice-Hall: Englewood Cliffs, NJ, 1981.
25. Bowen, C.H.; Huang W.C. “Application of reliability and economic
tics of the waste feedstock and the end product property analyses to the selection of a refuse shredding system,” in Proceedings
required by the recycling sectors or subsequent waste- of 1978 National Waste Processing Conference; ASME: New York, 1978.
26. Sommer, E.J.; Kenny, G.R.; Roos, C.E. J. Air & Waste Manage. Assoc.
to-energy facilities. More information on the sources of 1989, 39(4), 511-516.
27. Jackson, D.V. Resources and Conservation 1987, 14, 1-13.
the waste would be helpful, such as the percentage from 28. Bunsow, W.; Dobberstein, J. Resources and Conservation 1987, 14, 249-256.
residential and commercial sources and whether there is 29. Wilson, D.C. Resource Recovery and Conservation 1979, 4, 161-188.
30. Buckley, T.J.; Domalski, E.S. “Evaluation of data on higher heating
any separation being done by the residents. values and elemental analysis for refuse-derived fuels,” in Proceedings
of 1988 National Waste Processing Conference; ASME: New York, 1988.
31. Kirklin, D.R.; Colbert, J.C.; Decker, P.H.; Ledford, A.E.; Ryan, R.V.;
ACKNOWLEDGMENT Domalski, E.S. Resources and Conservation 1988, 9, 281-300.
32. Wilkins, E.S. J. Env. Sci. & Health:Part A 1983, 6(18), 747-772.
The authors acknowledge support from the National Sci- 33. Buekens, A.G.; Schoeters, J.G. Conservation & Recycling 1986, 9(3),
ence Council (NSC-85-2621-P-006-033) and helpful com- 253-269.

ments of anonymous reviewers in this study.

About the Authors

REFERENCES Ni-Bin Chang (corresponding author) is a professor in the
1. 1990-1994 Yearbook of Statistics of Environmental Protection in Taiwan;
Environmental Protection Administration. Office of Statistics: Taipei, Department of Environmental Engineering, National Cheng-
Taiwan, R.O.C. Kung University. Ying-Hsi Chang and W. C. Chen are gradu-
2. Spencer, D.B.; Temple, J. W. “Large scale rotary shredder performance ate students in the same Department. Ni-Bin Chang can
testing,” in Proceedings of 1984 National Waste Processing Conference;
American Society of Mechanical Engineers (ASME): New York, 1984. be reached by phone or fax at 011-886-6-2342357 or e-
3. Trezek, G.J. Significance of Size Reduction in Solid Waste Management; mail at T15579@[Link].
U.S. Environmental Protection Agency: Washington, DC, 1977; EPA-
600/2-77-131.

544 Journal of the Air & Waste Management Association Volume 48 June 1998