j ourna l homepage: www.e lsev ie r .com/ locate / fuproc
Production, characterization and fuel properties of alternative diesel fuelfrom pyrolysis of waste plastic grocery bags☆
Brajendra K. Sharma a,⁎, Bryan R. Moser b, Karl E. Vermillion b, Kenneth M. Doll b, Nandakishore Rajagopalan a
a Illinois Sustainable Technology Center, Prairie Research Institute, University of Illinois, Urbana-Champaign, 1 Hazelwood Dr., Champaign, IL 61820, USAb United States Department of Agriculture, Agricultural Research Service, National Center for Agricultural Utilization Research, Peoria, IL 61604, USA
☆ Disclaimer:Mention of tradenames or commercial proly for the purpose of providing specific information and dor endorsement by the U.S. Department of Agriculture. USvider and employer.⁎ Corresponding author.
Pyrolysis of HDPE waste grocery bags followed by distillation resulted in a liquid hydrocarbon mixture withaverage structure consisting of saturated aliphatic paraffinic hydrogens (96.8%), aliphatic olefinic hydrogens(2.6%) and aromatic hydrogens (0.6%) that corresponded to the boiling range of conventional petroleum dieselfuel (#1 diesel 190–290 °C and #2 diesel 290–340 °C). Characterization of the liquid hydrocarbon mixture wasaccomplished with gas chromatography–mass spectroscopy, infrared and nuclear magnetic resonance spectros-copies, size exclusion chromatography, and simulated distillation. No oxygenated species such as carboxylicacids, aldehydes, ethers, ketones, or alcohols were detected. Comparison of the fuel properties to the petrodieselfuel standards ASTM D975 and EN 590 revealed that the synthetic product was within all specifications afteraddition of antioxidants with the exception of density (802 kg/m3). Notably, the derived cetane number (73.4)and lubricity (198 μm, 60 °C, ASTM D6890) represented significant enhancements over those of conventionalpetroleum diesel fuel. Other fuel properties included a kinematic viscosity (40 °C) of 2.96 mm2/s, cloud pointof 4.7 °C, flash point of 81.5 °C, and energy content of 46.16 MJ/kg. In summary, liquid hydrocarbons withappropriate boiling range produced from pyrolysis of waste plastic appear suitable as blend components forconventional petroleum diesel fuel.
Plastic retail bags are ubiquitous in modern society because they rep-resent a convenient means to transport purchased goods from the super-market to the home. Plastic bags are plentiful, inexpensive to produce,sturdy yet low weight, and easy to store and transport. However, thesame properties that make them commercially successful also contributeto their proliferation in the environment. Although they are recyclable,the U.S. EPA noted that only 13% of the approximately one trillion pro-duced in 2009were recycled in the U.S. [1]. The remainderwere disposedof in landfills, released into the environment as litter, or used in secondaryapplications by end-users eventually ending in landfills. Plastic bags maytake centuries to naturally decompose due to their stable chemical com-position [2]. In addition to being a source of litter in urbanized areas, plas-tic bags exacerbate localized flooding by clogging municipal drainagesystems and constitute a significant portion offloating anthropogenicma-rine debris [3–5]. In fact, plastic bags contribute to the so-called Great Pa-cific Garbage Patch of floating refuse in the Pacific Ocean and have been
ducts in this publication is sole-oes not imply recommendationDA is an equal opportunity pro-
ghts reserved.
detected as far north and south as the poles [3–5]. Once in the environ-ment, plastic bags are lethal to animals that ingest or become entangledin them [6–8]. Because of these and other factors, various regional andna-tional governments have banned or are contemplating bans or fees onplastic bags [9].
Standard plastic bags consist of thin polyethylene (PE) sheetsproduced commercially from polymerization of ethylene. PE is dividedinto categories based on density and molecular branching frequency.The two types most important to production of plastic bags are low-density PE (LDPE) and high-density PE (HDPE). HDPE is a copolymerwith up to 1% 1-butene and ismade historically with either Cr or Zieglercatalysts at 1–16 MPa at temperatures as low as 60 °C. More recently,single site catalysts such as metallocenes have been employed [10].LDPE is produced at high temperatures (200–300 °C) and supercriticalethylene pressures (130–260 MPa) using peroxide-free radical initia-tors [10]. HDPE is a linear copolymer with a density range of 0.945–0.965 g/cm3 whereas LDPE is branched with densities ranging from0.915 to 0.925 g/cm3 [10]. Due to these differences in structure, the crys-talline melting point, softening point and tensile strength of LDPE areconsiderably lower than the corresponding values for HDPE [10]. How-ever, LDPE shows higher elongation at break andhigher impact strengththan does the more rigid HDPE [10]. It is also translucent rather thanopaque due its lower crystallinity (55%) relative to HDPE (85–95%)[10]. HDPE is more commonly utilized for production of plastic bagsdue to its greater tensile strength coupled with its less energy-
a Values in parentheses represent standard deviations from the reported means (n = 3). For flash point, n = 1.b For No. 2 grade S15 (15 ppm S) ULSD.c Not specified.d With 1000 ppm BHT added.
Plasticcrude oil
74%
Solid residue17%
Gases9%
Product yield from pyrolysis of plastic grocery bags
Fig. 1. Product yield from pyrolysis of plastic grocery bags.
intensive production process. Pyrolysis is defined as the irreversible an-aerobic thermochemical decomposition of material at elevated temper-ature (300+ °C). The principal benefit of pyrolysis is conversion of lowenergy density substrates into higher density liquid (bio-oil) and solid(biochar) fractions. A low-density volatile (syngas) fraction is alsoproduced. Pyrolysis has been utilized for millennia to produce charcoaland coal. More recently, pyrolysis is used to produce charcoal, activatedcarbon, coke, carbon fiber, and methanol, among others. The distribu-tion of products (bio-oil, biochar and syngas) is dependent on thetype of pyrolysis, reaction conditions and feedstock. Pyrolysis is classi-fied into four categories: slow, fast, flash, and gasification. Of these,fast and flash pyrolysis maximizes bio-oil production, slow pyrolysisaugments the yield of biochar and gasification maximizes syngas pro-duction. With regard to production of liquid transportation fuels, fastor flash pyrolysis is employed to produce bio-oil [11–14]. The propertiesand composition of bio-oil such as highmoisture andheteroatomcontent,presence of oxygenates such as organic acids, and broad distillation curveprevent its direct use as a transportation fuel; thus upgrading such ashydroprocessing and distillation is necessary [14–16].
Fast or flash pyrolysis has been reported on biological materials suchas wood [13], triglycerides [17], grasses [18], shrubs [19], corn cobs andstover [20], alfalfa [21], oilseed presscakes [22], and pig compost [23],among others. Examples of fast pyrolysis on non-biological feedstocks in-clude scrap tires [24,25], sewage sludge [25], general municipal solidwaste [26], waste electrical and electronic equipment [27], and variousplastics [25,28–36]. The plastics include polystyrene [30–32], poly(vinylchloride) [30,31], polypropylene [31–34], PE terephthalate [32], acryloni-trile–butadiene–styrene [32], and PE [30–32,35,36]. In some cases plasticswere co-pyrolyzed with other materials such as waste motor oil [32].With regard to fast pyrolysis of PE, pyrolysis of LDPE [30], HDPE [35,36]and various mixtures [31,32] was reported. In all PE studies, the proper-ties of the resulting bio-oils were not reported, nor were the upgradingto fuel-grade hydrocarbons and subsequent fuel property determination.
The objective of our study was the production, characterization andevaluation of alternative diesel fuel from pyrolysis of HDPE waste gro-cery bags. Comparison of our pyrolyzed polyethylene hydrocarbons(PPEH) with conventional petroleum-derived ultra-low sulfur(b15 ppm S) diesel (ULSD) fuel was a further objective, along witha comparison to petrodiesel standards such as ASTM D975 and EN590 (Table 1). Blends of PPEH with ULSD and biodiesel were pre-pared and the resultant fuel properties measured. It is anticipatedthat these results will further understanding of the applicability
and limitations of HDPE as a feedstock for the production of alterna-tive diesel fuel.
2. Materials and methods
2.1. Materials
Plastic HDPE grocery bags were collected from local retailers andrepresent the typical ones used in grocery stores. Summer grade ULSDwas donated by a major petrochemical company. With the exceptionof conductivity and corrosion inhibitor additives, ULSD contained noperformance-enhancing additives. Soybean oil methyl esters (SME)were donated by a BQ-9000 certified commercial producer. All otherchemicals were obtained from Sigma-Aldrich Corp (St. Louis, MO). Allmaterials were used as received.
2.2. Pyrolysis of HDPE to produce plastic crude oil
Thermochemical conversion of plastic grocery bags (HDPE) to oilswere conducted using a pyrolysis batch reactor in triplicate. Pyrolysiswas performed in a Be-h desktop plastic to oil system (E-N-Ergy, LLC,Mercer Island, WA) containing a 2 L reactor and oil collection systemusing approximately 500 g of plastic grocery bags each time. The pyrol-ysis reactor has two heating zones (upper and lower); the upper and
0% 10% 20% 30% 40% 50%
Fraction 1 (<190°C)
Fraction 2 (190-290°C)
Fraction 3 (290-340°C)
Fraction 4 (>340°C)
Average yield, %
Fig. 2. Distillate fractions yield (%) on distillation.
Table 3Elemental analysis of waste plastic grocery bags and plastic oil distillates. (The numbersshown are average of two measurements, while oxygen is determined by difference).
lower temperatures were set to 420 and 440 °C, respectively. Once thereactor reached the set temperatures, a reaction time of 2 h wasemployed from that point on. Vapors produced as a result of pyrolysiswere condensed over water as plastic crude oil (PCO). The upper oillayer was separated and weighed. The reactor lid was opened oncethe temperature was below 50 °C to remove the remaining residualsolid material and weighed separately. The mass balance yields werecalculated as the ratio of the corresponding product phase (liquid andsolid) obtained in 12 batch experiments to the initial feedstock mass.Lastly, the gas-phase yields were calculated based on the resultingmass difference.
2.3. Distillation of plastic crude oil to yield diesel-range hydrocarbons
Distillation of PCO was performed in a Be-h desktop plastic to oilsystem. A known amount of PCO (1 L) was added to the Be-h reactorvessel. The oil collection tank was cleaned by removing the water anddried before starting distillation. For collecting the gasoline equivalentfraction (b190 °C), the upper and lower temperatures were set to 175and 190 °C, respectively. Once the liquid stopped dripping into thecollection vessel, the gasoline equivalent fractionwas removed, filtered,and weighed to provide yield. The upper zone temperature was thenraised to 275 °C and lower zone to 290 °C to collect a #1 dieselequivalent fraction (190–290 °C). The # 2 diesel equivalent fraction(290–340 °C)was then collected by setting the upper zone temperatureto 330 °C and lower zone to 340 °C. Thematerial remaining in the reac-tor vessel was an atmospheric residue equivalent fraction (N340 °C),whichwas removed using a siphon pump once the reactor temperaturewas below 50 °C. All fractions except the atmospheric residue equiva-lent (N340 °C) were filtered through Whatman filter paper #4 toremove residual solid particles.
2.4. Chemical characterization of plastic oil fractions
Elemental analysis of HDPE plastic grocery bags and PCO fractionswas conducted at the University of Illinois Microanalysis Laboratory(Urbana, IL). Samples were processed for total CHN (carbon/hydro-gen/nitrogen) using an Exeter Analytical (Chelmsford, MA) CE-440Elemental Analyzer. Oxygen was calculated by mass balance closure.
Table 2Distillate fraction yields (wt.%) of plastic crude oils (PCO) using simulated distillation (HTGC–F
2.4.1. Gas chromatography–mass spectroscopy (GC–MS)Samples were analyzed for composition utilizing an Agilent (Palo
Alto, CA) model 7890 GC equipped with a model 7683 auto-injectorand a model 5975 MSD in electron impact (EI) mode. An Agilent DB-35MS column (30 m × 0.320 mm; 0.25 μm film thickness) wasused with a helium flow rate of 0.509 mL/min. The temperature pro-gram began with a hold at 30 °C for 10 min followed by an increase at1 °C/min to 195 °C, then 35 °C/min to 330 °C, which was held for1 min. The injector and column transfer line heater were both set to340 °C. The detector inlet and MS quadropole temperatures were 220and 150 °C, respectively. The injection volume was 1 μL with a splitratio of 50:1. Samples (5 mg) were dissolved in heptane (1 mL). Asecond set of samples was dissolved in octadecene so that compoundsof similar or shorter retention time to heptane could also be identifiedand quantified. Identity of chromatographic peaks was establishedfrom MS data, which was matched to the Agilent spectral library. Themolecular ion was observed in all cases, thus providing further corrob-oration of peak identity.
2.4.2. Simulated distillation by GC–FIDThe boiling point distribution of PCO fractions was obtained
by performing simulated distillations according to ASTM 7169.The analysis was performed on 1% (w/w) sample solutions indichloromethane using an HP 5890 Series II FID gas chromatographequipped with a temperature programmed vaporizer injector, HP7673 autosampler, and a Durabond DB-HT-SimDis column by AgilentJ&WScientific (5m× 0.53mm id, 0.15 μm film) as described by Vardonet al. [37–39].
2.4.3. Size exclusion chromatography (SEC) analysisMolecular weight (MW) distributions were determined by SEC as
reported by Vardon et al. [37–39] on a Waters (Milford, MA) StyragelHR1 SEC column (7.8 mm × 300 mm), Waters 2414 RI detector, and
THF asmobile phase (1.0 mL/min). The resulting chromatographic datawas processed using Matlab (Natick, MA) software to provide theweight-average MW (Mw) and polydispersity index (PDI).
2.4.4. NMR and FT-IR spectroscopyChemical functionality information was obtained by analyzing the
fractions using Fourier-Transform infrared (FT-IR) and nuclear magneticresonance (NMR) spectroscopies.
1H and 13C NMR data were collected using a Bruker Avance-500spectrometer (Billerica, MA) running Topspin 1.4 pl8 software operat-ing at 500 MHz (125 MHz for 13C NMR) using a 5-mm BBO probe.Samples were dissolved in CDCl3 (Cambridge Isotope Laboratories,Andover, MA) and all spectra were acquired at 26.9 °C. Chemical shifts(δ) are reported as parts per million (ppm) from tetramethylsilanebased on the lock solvent.
FT-IR spectra were obtained on a Thermo-Nicolet Nexus 470 FT-IRspectrometer (Madison, WI) with a Smart ARK accessory containing a45 ZeSe trough in a scanning range of 650–4000 cm−1 for 64 scans ata spectral resolution of 4 cm−1.
2.5. Fuel properties
Properties were measured (n = 3) following AOCS, ASTM, and CENstandard test methods using instrumentation described previously[40–42] and reported in Table 1: acid value (AV, mg KOH/g), AOCS Cd3d-63; cloud point (CP, °C), ASTM D5773; cold filter plugging point(CFPP, °C), ASTM D6371; density (g/cm3), ASTM D4052; flash point(FP, °C), ASTM D93; gross heat of combustion (higher heating value,HHV, MJ/kg), ASTM D4809; induction period (IP, h), EN 15751; mois-ture content (ppm), ASTM D6304; kinematic viscosity (KV, mm2/s),ASTM D445; lubricity (μm), ASTM D6079; pour point (PP, °C), ASTMD5949; specific gravity (SG), ASTM D4052; and sulfur (S, ppm), ASTMD5453. For a greater degree of precision, PP was measured with a reso-lution of 1 °C instead of the specified 3 °C increment. Derived cetanenumber (DCN) was determined (n= 32) by Southwest Research Insti-tute (San Antonio, TX) following ASTMD6890. Oxidation onset temper-ature (OT, °C) was measured by pressurized differential scanningcalorimetry (PDSC) following the procedure outlined in [43]. Surfacetension (mN/m) was measured as described by Doll et al. [44].
2.6. Preparation of PPEH–petrodiesel blends
Blends were prepared at room temperature (22–24 °C) by pipettingprecisely measured volumes followed by agitation of the contents to
ensure homogeneity. Properties of PPEH were measured at the 10, 20,30, 40, and 50 volume percent (vol.%) levels in ULSD. These sampleswere thus labeled as P10, P20, P30, P40 and P50 as analogous to themore commonly encountered BXX notation utilized when describingblends of biodiesel with petrodiesel.
3. Results and discussion
3.1. Preparation and chemical composition of pyrolyzed plastics
The pyrolysis temperature range of 420–440 °C was chosen basedon previous studies [34]. These temperatures resulted in decompositionreactions of HDPE to provide hydrocarbons of different chain lengths.Pyrolysis of waste plastic grocery bags at temperatures of 420–440 °Cprovided 74% yield of liquid product referred to as PCO, as shown inFig. 1. Although not determined in the present paper, literature datasuggested gaseous product obtained from pyrolysis of PE consistedprimarily of ethane and ethene (C2, 52%) and C4 (32%) compounds[34]. The higher solid residue yield (17%) is likely due to the inorganiccontent and/or char content and/or unconverted HDPE. As the pyrolysisof PE has higher activation energy (280–320 kJ/mol) compared to poly-propylene (190–220 kJ/mol), therefore, increasing the pyrolysis tem-perature to certain extent could result in increased amounts of theliquid fraction [34]. Also, this residuemay have been the fraction boilingabove 420 °C (analogous to the higher boiling vacuum gas oil fraction,VGO frompetroleumdistillation). Further thermal cracking of this prod-uct could have been achieved by increasing pyrolysis temperature and/or time, which we speculate would have resulted in higher yields of thedesired PCO fraction. This residue alongwith the VGO fraction from PCOhave potential to be used as lubricant basestocks, which upon furtherrefining such as dewaxing/wax isomerization may yield API Group II/III lubricant base oils.
The PCO thus obtained after pyrolysis of waste plastic grocery bagswas distilled into four fractions (b190; 190–290; 290–340; and 340+°C equivalent of motor gasoline (MG), diesel#1 (PPEH-L), diesel #2(PPEH-H) and VGO respectively. The product yields are represented inFig. 2. Similar results were obtained from SimDist analysis of PCOswith maximum coefficient of variation ranging from 0 to 7% (Table 2).In the absence of a catalyst, the major product is PPEH-L (41%). Theproduct distribution can be changed with the use of zeolite catalystssuch as ZSM-5, which will increase conversion to more low boilingproducts, such as MG and PPEH-L [34].
Elemental analysis of waste plastic grocery bags and PCO fractionsrevealed less than 0.5% nitrogen content and less than 0.7% oxygen con-tent (Table 3). As expected,waste plastic grocery bags have an empiricalformula of CH2.1N0.005O0.007 quite similar to that of polyethylene (CH2).Higher carbon and hydrogen content and lower oxygen and nitrogencontent resulted in a higher calculated HHV [37–39] of 49–50 MJ/kgfor most of the fractions, making these high energy liquid fuels(Table 3). The calculated values were slightly higher than the actualdeterminations (Table 1).
Analysis by SEC was employed to determine the MW distribution ofconstituents in PPEH-L, PPEH-H and ULSD. The weight-average MW(Mw), PDI, and MW at peak end were determined from a retentiontime calibration curve and signal intensities from SEC data of the PCOfractions. Corresponding to the lower temperature range collected dur-ing distillation for PPEH-L (190–290 °C) relative to PPEH-H(290–340 °C), PPEH-L displayed a lowerMW than PPEH-H. Such resultsindicated that, on average, theMWof constituents in PPEH-L was lowerthan that of PPEH-H (Table 4). Comparison to the MW obtained forULSD (Mw of 131withmaxMWof 299 g/mol) revealed greater similar-ity to PPEH-L than PPEH-H. For most of the PCO fractions polydispersityindex (PDI) ranged between 1.33 and 1.65, thus indicating a narrowdis-tribution of MWs for these fractions compared to the ULSD exhibiting ahigher PDI (2.08). This resulted in higher PDI of PPEH–ULSD blendscompared to PCO fraction (PPEH-H). As expected, upon increasing
Fig. 3. Simulated distillation of a plastic crude and its four fractions.
amounts of PPEH-H in ULSD, the Mw increased due to the higher Mw ofPPEH-H (212) relative to ULSD (131).
The boiling point distribution of PCO fractions was obtained usinghigh temperature GC–FID. Table 2 shows that the method developedfor boiling point distribution was repeatable with a CV of less than 7%for distribution of various fractions in PCO andwas similar to actual dis-tillation data. All PCOs contained a large percentage of fraction 2 (PPEH-L) followed by PPEH-H and MG. Around 98% of the PCO was distilledunder 400 °C, which is a good range for producing various fuels suchas naphtha, gasoline, aviation fuel, diesel, and fuel oil. The boilingpoint distribution of PCO and its four fractions is shown in Fig. 3. Asboiling point and MW distribution of PCO were similar to petroleumfractions and contained negligible heteroatom content, therefore, wespeculate that these PCOs will be compatible with petroleum crude oilfor refining in a conventional refinery. The compatibility is further
depicted in 10–50% blends of PPEH in ULSD as shown in Fig. 4. As thePPEH content increased in ULSD, the boiling point distribution shiftedtowards the higher boiling range, although overall the mixtureremained within the boiling range of diesel fuel.
Compositional analysis by NMR spectroscopy revealed the presenceof aromatic, olefinic and paraffinic protons in PPEH-L and PPEH-H. Asseen in Table 5, aromatics comprised 1.0 and 0.6% of the overall protoncontent of PPEH-L and PPEH-H, respectively. Unsaturated protons(aliphatic olefins) constituted 5.4% and 2.6% of PPEH-L and PPEH-H, re-spectively. The remainder of the content of PPEH-L (94.0%) and PPEH-H(96.8%) was composed of aliphatic saturated hydrocarbons. No oxygen-ated species such as carboxylic acids, aldehydes, ketones, ethers, or alco-hols were detected by either 1H or 13C NMR spectroscopies. Resultsobtained from FT-IR analysis, Fig. 5, confirmed those collected byNMR. The FT-IR spectrum was dominated mainly by aliphatic peaks,
Fig. 4. Simulated distillation of SG-ULSD and its blends with PPEH.
consistentwithhigh energy density, and included strongC–H stretching(3000–2800 cm−1) and C–H bending (1465 cm−1, 1375 cm−1) signals.
The principal constituents in PPEH-L and PPEH-Hwere quantified byGC–MS, as depicted by a representative chromatogram (Fig. 6). Inagreement with the NMR results, both samples were comprised of a
series of saturated and unsaturated hydrocarbons. The results, present-ed in Table 6, demonstrated that both fuels were composed of amixtureof hydrocarbons. However, PPEH-H contained heavier constituents, inagreement with SEC data and depicted in Fig. 6. GC–MS results furtherrevealed that PPEH-L contained 43.6% (peak area) of compounds
Table 5Relative percentages of aromatic, olefinic and paraffinic protons in PPEH-L, PPEH-H andULSD as determined by 1H NMR spectroscopy.a,b
a Paraffins = aliphatic alkanes; olefins = aliphatic alkenes.b Percentages determined using integration values obtained from 1H NMR spectra of
signals corresponding to chemical shifts indicative of the functionalities indicated(aromatics: δ 6.7–8.0 ppm; olefins: 4.5–6.0 ppm; saturates: 0.5–3.0 ppm).
containing one or more double bonds (olefins), while PPEH-Hcontained 28.2% of such olefin peaks. It was demonstrated earlier byothers as well that the total olefin content of light oil and gasoline wasaround 47% [34]. In PPEH-L, all of the components contained less than20 carbons, and molecules as small as octane were observed, whichwas similar to petroleum diesel typically containing between 8 and 21carbon atoms per molecule. Isomers in the range of C-11 to C-13 werethe most common in PPEH-L. The PPEH-H sample contained less ofthese smaller chains, with the highest concentration in the range of C-16 to C-18 isomers. Alkanes of up to 23 carbons were observed in thisfuel. (Fig. 7.)
3.2. Properties of PPEH and comparison to ULSD
To demonstrate that PCOs obtained by pyrolysis have utilization inexisting refineries, diesel like fuels (PPEH-L and PPEH-H) obtainedfrom PCO were compared with ULSD to demonstrate their applicabilityand limitations.
Depicted in Table 1 are fuel properties of PPEH-L, PPEH-H, a 1:1blend of PPEH-L with PPEH-H, and ULSD along with a comparison tothe petrodiesel standards ASTM D975 and EN 590. PPEH-L exhibitedexceptional CP, CFPP and PP values of −30.1, −31.0 and −37.3 °C,respectively. Such properties represented a significant enhancementover ULSD, which provided CP, CFPP and PP values of −17.5, −16.0and −20.3 °C, respectively. However, PPEH-H, with its greater contentof higher-melting longer-chain paraffinic constituents relative toPPEH-L and ULSD (Table 6), provided CP (4.7 °C), CFPP (3.7 °C) andPP (4.0 °C). The 1:1 PPEH-L/H blend, representative of summer gradeULSD, yielded cold flow properties (CP −5.9 °C, CFPP −6.0 °C, and PP−8.3 °C) intermediate to those of the neat materials, but not as favor-able as ULSD.
1500 2000 2500 3000
cm-1
Fig. 5. The FT-IR spectra of PPEH-H (top), PPEH-L (middle), compared to ULSD (bottom),showing pure hydrocarbon content of all three fuels.
EN 590 prescribes a minimum oxidative stability (IP, 110 °C) of20 h whereas ASTM D975 contains no such specification. BothPPEH-L (3.9 h) and PPEH-H (12.9 h) did not meet the threshold pre-scribed in EN 590. The 1:1 PPEH-L/H blend also was below the min-imum limit (7.7 h). The IP of ULSD (N24 h) conformed to the limitspecified in EN 590. The presence of unsaturated constituents(Table 5) was speculated as the reason for the reduced stabilities ofPPEH-L and PPEH-H versus ULSD. The lower content of olefinic com-pounds in PPEH-H (28.2%) was postulated as the cause for its en-hanced stability relative to PPEH-L (43.6%). Employment ofantioxidants improved the stabilities of PPEH-L, PPEH-H and their1:1 mixture. Specifically, 1000 ppm BHT yielded IPs N 24 h forPPEH-H and the 1:1 mixture, thus exceeding the threshold specifiedin EN 590. However, PPEH-L treated with BHT was below the mini-mum limit (14.4 h), which indicated that a hydrogenation step maybe required before its use as a blend component in ULSD.
Both the American and European petrodiesel standards containspecifications for KV at 40 °C with ASTM D975, prescribing a range of1.9–4.1 mm2/s and EN 590 specifying a range of 2.0–4.5 mm2/s. TheKV of PPEH-L (1.20 mm2/s) was below the minimum limits due to itshigh content of shorter-chain, lower MW constituents. Both PPEH-H(2.96 mm2/s) and ULSD (2.28 mm2/s) exhibited KVs within the rangesspecified in the standards. The KV of the 1:1 PPEH-L/H mixture(2.08mm2/s)was in between the values obtained for the neatmaterialsand conformed to the limits specified in ASTM D975 and EN 590.
ASTMD975 and EN 590 specifyminimum limits for CN of 40 and 51,respectively. PPEH-L (54.6) and PPEH-H (73.4) exhibited DCNsmeetingthe limits specified in the standards and are higher than that obtainedfor ULSD (47.4). An earlier study reported cetane numbers in therange of 62–69 for light oils prepared from pyrolysis of HDPE [34]. Thehigher DCN of PPEH-H versus PPEH-L was attributed to its highercontent of longer-chain paraffins, as longer chains result in higherDCNs [45]. A DCN of 66.3 was observed for the 1:1 PPEH-L/H mixture,whichwas in between the neat materials and above the limits specifiedin ASTM D975 and EN 590.
Theminimum limits prescribed for FP in ASTMD975 and EN 590 are52 and 55 °C, respectively. Due to its comparatively high content ofshorter-chain, low MW constituents, PPEH-L exhibited a FP (b30 °C)significantly below the minimum limits specified in the petrodieselstandards, and also considerably lower than the values obtained forPPEH-H (81.5 °C) and ULSD (65.0 °C). Accordingly, the 1:1 PPEH-L/Hmixture exhibited a FP of 90.3 °C and was within the limits prescribedin ASTM D975 and EN 590.
Maximumwear scars of 520 and 460 μmare specified as upper limitsfor lubricity (60 °C) in ASTMD975 and EN 590. All of the PPEH samplesprovided wear scars significantly below the thresholds listed in thepetrodiesel standards. The longer wear scar noted for PPEH-L(293 μm) versus PPEH-H (198 μm) was speculated to be due to thehigher content of longer-chain constituents in PPEH-H, as increasingchain length results in lower wear and consequently shorter wear scarlengths [46]. The wear scar obtained for ULSD (581 μm) was above theupper limits prescribed in the standards and significantly longer thanobserved for the PPEH samples.
Sulfur, which poisons vehicle emissions control devices and contrib-utes to environmental pollution, is limited tomaximum levels of 15 and10 ppm in ASTM D975 and EN 590. The concentrations of sulfur in allsamples, including ULSD, were below 10 ppm. Similarly, the contentof moisture in all samples was below the maximum limit of 200 ppmprescribed in EN 590. ASTM D975 does not contain a moisturespecification.
ASTMD975 does not contain limits on density, but EN590 specifies adensity of 820–845 kg/m3 at 15 °C. Neither PPEH-L (776 kg/m3) norPPEH-H (802 kg/m3) provided densities meeting the range specifiedin EN 590. Accordingly, the density of the 1:1 PPEH-L/H blend(791 kg/m3) was below the minimum limit. ULSD conformed to EN590 with a density of 849 kg/m3. The lower densities of the PPEH
Fig. 6. The GC–MS chromatogram of PPEH-H showing the separation of the alkane and alkene components of the mixture over the first 170 min of run-time.
samples versus ULSDmay be attributed to the greater percentage of ar-omatics in ULSD. Aromatics exhibit higher densities than linear,branched and cyclic hydrocarbons more commonly encountered inthe PPEH samples [47]. The higher density of PPEH-H versus PPEH-Lmay be attributed to the higher content of longer-chain constituents
Table 6Principal constituents (area %) identified by GC-MS in PPEH-L and PPEH-H.
in PPEH-H, as these tend to have higher densities than shorter-chain hy-drocarbons. Also measured was SG, which is not specified in eitherASTM D975 or EN 590. As was the case with density and for essentiallythe same reasons, the SGs of PPEH-L (0.777) and PPEH-H (0.803) werelower than observed for ULSD (0.841). Correspondingly, the SG of the1:1 PPEH-L/H mixture (0.792) was in between the neat materials andlower than ULSD.
Although ST is not specified in either ASTM D975 or EN 590, itnevertheless influences fuel atomization in combustion chambers ofdiesel engines [48]. The STs of PPEH-L (22.5 mN/m) and PPEH-H (24.7(mN/m) at 40 °C were below ULSD (25.1 mN/m). It is speculated thatthe higher content of longer-chain constituents in PPEH-H was respon-sible for its higher ST versus PPEH-L, as a positive correlation betweenchain length and ST was established previously [44]. Consequently,the ST of the 1:1 PPEH-L/H blend (23.6 mN/m) was in between thevalues obtained for the neat materials and lower than ULSD.
TheHHVs of PPEH-L (45.86 MJ/kg) and PPEH-H (46.16 MJ/kg) alongwith the 1:1 blend (46.04 MJ/kg)were higher thanULSD (45.15 MJ/kg).It was no surprise that the HHV of PPEH-H was greater than PPEH-L, aslarger hydrocarbons generally contain more energy content. The higherHHVs of the PPEH samples relative to ULSDwas attributed to the highercontent of aromatics in ULSD, as aromatics contain less energy than sat-urated constituents found in higher abundance in PPEH. Energy contentis not specified in the petrodiesel standards.
Thediscussion above demonstrated that PPEH-L, PPEH-H, and PPEH-L/H had good properties for further fuel-like utilization similar to petro-leum diesel.
3.3. Properties of PPEH blended with ULSD
Depicted in Tables 7 and 8 are fuel properties of PPEH-L (Table 7)and PPEH-H (Table 8) blended with ULSD. For each sample, blends of10 (P10), 20 (P20), 30 (P30), 40 (P40) and 50 (P50) vol.% in ULSDwere investigated. With regard to cold flow properties, as the percent-age of PPEH-L increased in blends with ULSD, values for CP, CFPP andPP became progressively lower due to the superior low temperature
0
5
10
15
5 10 15 20 25
PPEH-LPPEH-H
Ch
rom
ato
gra
m a
rea
(%)
Carbon number
Fig. 7. The composition of PPEH-H and PPEH-L vs. carbon number.
performance of PPEH-L relative to ULSD. In the case of PPEH-H, the op-posite trend was elucidated in which cold flow properties (CP and PP)deteriorated as the concentration of PPEH-H in ULSD increased. CFPPof PPEH-H was not measured due to insufficient sample. Similarly, oxi-dative stability decreased as the percentage of PPEH increased in blendswith ULSD. Comparison to the IP specification in EN 590 revealed thatonly the P10–30 blends of PPEH-H were above the minimum specifica-tion of 20 h. Results obtained from measurement of OT corroboratedthose obtained for IP: deterioration of stability as the concentration ofPPEH increased in ULSD as indicated by progressively lower OTs.
Because the order of KV of neat materials was PPEH-L (lowest)b ULSD b PPEH-H (highest), opposite trends were noticed whenPPEH-L and PPEH-H were blended with ULSD. Specifically, lower KVswere noted as the concentration of PPEH-L increased in blends withULSD whereas higher KVs were observed with progressively higherconcentrations of PPEH-H. Comparison to fuel standards listed inTable 1 revealed that all PPEH-H blends along with the P10 PPEH-L/ULSD were within the limits prescribed in the petrodiesel standards.However, the P30–P50 PPEH-L/ULSD samples did not conform to thepetrodiesel standards. The P20 PPEH-L/ULSD sample was satisfactorywhen compared against ASTM D975 but not EN 590.
As the concentration of PPEH-L and PPEH-H in ULSD increased,progressively higher DCNs were obtained. All blends met theminimumlimit of 40 prescribed in ASTM D975, but only the P20–P50 blends ofPPEH-H and the P50 blend of PPEH-L satisfied the more stringent
Table 7Fuel properties of pyrolyzed polyethylene hydrocarbons (PPEH-L) blended with ULSD.a
Units P10 P20
Low temperature °CCP −15.9 (0.1) −16.8 (0PP −25.3 (0.6) −27.7 (0CFPP −17.7 (0.6) −19.7 (0Oxidative stability:IP, 110 °C h 16.5 (1.6) 10.0 (0.2)OT °C 194.9 (0.1) 190.8 (0.8KV, 40 °C mm2/s 2.11 (0) 1.96 (0)DCN 48.4 (0.9) 48.9 (0.9)Flash point °C 54.0 (0.7) 48.5 (0.7)Wear scar, 60 °C μm 538 (5) 512 (4)Sulfur ppm 8 8SG, 15 °C 0.835 (0) 0.829 (0.0Density, 15 °C kg/m3 834 (1) 828 (0)Moisture ppm 55 (1) 56 (2)HHV MJ/kg 45.22 (0.10) 45.20 (0.0
a Values in parentheses represent standard deviations from the reported means (n = 3). Fo
threshold of 51 specified in EN 590. The low DCN of ULSD relative tothe PPEH samples was speculated as the reason for the low DCNs oflow-level blends of PPEH-L/H in ULSD.
Opposite trendswere notedwith regard to FP of PPEH-L and PPEH-Hblends in ULSD. Specifically, FP decreased as the percentage of PPEH-Lincreased in blends with ULSD whereas higher FPs were noted as thecontent of PPEH-H increased in blends. All of the PPEH-H blends werewithin the limits prescribed in ASTM D975 and EN 590 whereas noneof the PPEH-L blends met the limits set forth EN 590. The P10 PPEH-L/ULSD blend was above the minimum FP specified in ASTM D975. Thelower FPs of the PPEH-L blends relative to the PPEH-H was postulatedto be due to the significantly lower FP of neat PPEH-L versus neatPPEH-H.
Progressively shorter wear scarswere observed as the concentrationof PPEH increased due to the enhanced lubricities of PPEH-L and PPEH-H relative to ULSD. The effectwasmore pronounced in the case of PPEH-H, as it exhibited better lubricity than PPEH-L.With the exception of theP10 PPEH-L blend, the lubricities of all blendswere below themaximumlimit specified in ASTM D975. However, only the P20–P50 PPEH-H andP40–P50 PPEH-L blends were below the more stringent limit specifiedin EN 590.
Density as well as SG decreased as the percentage of PPEH increasedin blends with ULSD. The effect was greater for PPEH-L blends, as thedensity and SG of PPEH-L were lower than those of PPEH-H, which inturn were lower than the corresponding values for ULSD. All blendswith the exception of the P40 and P50 blends with PPEH-L provideddensities that fell within the range specified in EN 590.
All PPEH-L and PPEH-H blend samples were below the maximumallowable limits listed in the petrodiesel standards for sulfur content.Furthermore, all blend samples contained less than 80 ppm moisture,thus conforming to the upper limit of 200 ppm set forth in EN 590.Lastly, energy content of the blend samples did not vary significantlywith concentration of ULSD. For both PPEH-L and PPEH-H blends, HHVincreased slightly with increasing percentage of PPEH due to the higherenergy contents of PPEH-L and PPEH-H versus ULSD.
3.4. Influence of blending biodiesel with PPEH
Due to the emergence of biodiesel (fatty acidmethyl esters preparedfrom lipids) as a significant source of alternative diesel fuel and itsindustrial production as evidenced by separate ASTM (D6751) and EN(14214) standards governing its composition and properties, blends ofbiodiesel with PPEH were investigated herein [49,50]. Because soybeanoil is the principal feedstock for production of biodiesel in the U.S., thusresulting in SME, it was of interest to the current study [49]. Shown inTable 9 are B2 and B5 blends of SME in PPEH-L, PPEH-H and ULSD
along with neat SME (B100). Addition of SME to PPEH-H and ULSDresulted in lower IP and OT as the percentage of SME increased. Thestability of PPEH-Lwas unaffected upon SME addition due to the similarIPs of neat SME (4.6 h) and PPEH-L (3.9 h). As the concentration of SMEin PPEH-L and ULSD increased from B0 to B5, progressively higher CPand PP values were noted. In the case of PPEH-H, the opposite trendwas noticed. The reason for this behavior was due to the order of coldflow properties of the neat materials: PPEH-H (highest CP and PP)N SME NN ULSD NN PPEH-L (lowest CP and PP). The higher KV of neatSME (4.09mm2/s) relative to PPEH-L, PPEH-H andULSD resulted in pro-gressively higher KVs as the blend percentage of SME increased. Due tothe excellent lubricity of SME (152 μm), significant reductions in wearscar length were noticed as the percentage of SME increased in theblends. Comparison to the petrodiesel standards revealed that blendsof SMEwith PPEH-H and ULSDwerewithin the specifications for lubric-ity and KV. However, the PPEH-L blends satisfied the lubricity limits butdid not fall within the ranges specified for KV. It was found that dieselobtained from pyrolysis of plastic is as compatible with biodiesel asULSD due to quite similar hydrocarbon structures and chain length dis-tribution of molecules.
Depicted in Table 10 is the influence on cold flow properties of blend-ing 5, 10 and 20% PPEH-L, PPEH-H and ULSDwith SME alongwith a com-parison to neat SME. In the case of PPEH-H, CFPPwas not determined dueto lack of sample availability. Both PPEH-L and ULSDwere effective at de-pressing coldflowproperties of SME, although the influencewas linear asopposed to additive. PPEH-H,with its inferior coldflowproperties, causedCP and PP to increase as the percentage of PPEH-H was increased inblends with SME.
4. Conclusions
Pyrolysis of HDPE waste plastic grocery bags followed by distillationresulted in a major liquid hydrocarbon product (PPEH-L) with average
Table 9Influence of blending soybean oil methyl esters (SME) with PPEH-L and PPEH-H on fuel prope
SME PPEH-L
Units (B100) B2 B5
Oxidative stabilityIP, 110 °C h 4.6 (0.1) 3.9 (0.1) 3.9OT °C 175.4 (0.5) 174.8 (0.9) 176.5Low temperatureCP °C 0.3 (0.1) −28.5 (0.1) −26.8PP °C −1.7 (0.6) −36.7 (0.6) −35.7KV, 40 °C mm2/s 4.09 (0) 1.23 (0) 1.27Wear scar, 60 °C μm 152 (2) 267 (2) 201
a Values in parentheses represent standard deviations from the reported means (n = 3).
structure consisting primarily of saturated aliphatic paraffinic hydro-gens (94.0%) and smaller amounts of aliphatic olefinic hydrogens(5.4%) and aromatic hydrogens (1.0%) that corresponded to the boilingrange typical of conventional petroleum diesel fuel (190–290 °C).Negligible heteroatom-containing species were detected from elemen-tal analysis. Also obtained was a heavier boiling fraction (290–340 °C)equivalent of diesel#2 from distillation of the crude pyrolysis product,PPEH-H,which also consisted of paraffinic protons (96.8%), olefinic pro-tons (2.6%) and aromatic protons (0.6%). Based on the results obtainedafter determination of fuel properties and comparison to petrodieselstandards, the following conclusions were made regarding the applica-bility of these materials as alternative liquid transportation fuels:
1. PPEH-H is more appropriate as an alternative diesel fuel because itexhibited higher values for FP, IP, KV, DCN, HHV, density, and lubric-ity than PPEH-L.
2. PPEH-H, after addition of antioxidants, met all ASTM D975 and EN590 fuel specifications with the exception of density in the case ofEN 590.
3. PPEH-L did not meet EN 590 specifications for IP, KV, FP, and densitydue to its higher content of lower MW constituents.
4. A 1:1mixture of PPEH-H and PPEH-Lmet all ASTMD975 and EN 590specifications with the exception of density and IP in case of EN 590,therefore PPEH-L and PPEH-H distillates can be collected together toprovide ~64% diesel equivalent fraction from pyrolysis of plasticgrocery bags.
5. P10–P30 blends of PPEH-H with ULSD met all ASTM D975 specifica-tions whereas only the P20–P30 blends met all EN 590 limits. P40and P50 blends require antioxidants to meet the oxidative stabilityspecification listed in EN 590.
6. P10 blend of PPEH-L with ULSD met all ASTM D975 specificationsexcept lubricity, while none of the blends of PPEH-L with ULSD metEN 590 specifications primarily due to poor DCN, IP, FP, and KV.
Table 10Influence of blending PPEHwith SME on cold flow properties along with a comparison toblends in ULSD.a
Sample (vol.%) CP(°C)
CFPP(°C)
PP(°C)
SME 0.3 (0.1) −3.7 (0.6) −1.7 (0.6)5% PPEH-L in SME −0.6 (0) −5.7 (0.6) −3.0 (0)10% PPEH-L in SME −1.7 (0.1) −6.7 (0.6) −4.0 (0)20% PPEH-L in SME −3.9 (0.2) −8.7 (0.6) −6.7 (0.6)5% PPEH-H in SME 0.3 (0.1) N/Db −1.0 (0)10% PPEH-H in SME 0.4 (0.1) N/Db −1.0 (0)20% PPEH-H in SME 1.1 (0.1) N/Db 0.0 (0)5% ULSD in SME −0.3 (0.2) −4.3 (0.6) −2.3 (0.6)10% ULSD in SME −1.0 (0) −6.7 (0.6) −3.7 (0.6)20% ULSD in SME −2.6 (0.1) −8.7 (0.6) −5.0 (0)
a Values in parentheses represent standard deviations from the reported means(n = 3).
7. Diesel obtained frompyrolysis of plastic is as compatiblewith biodie-sel as ULSD due to similar hydrocarbon structures and chain lengthdistribution of molecules.
8. Biodiesel blends with PPEH-Hmet the specifications for lubricity andKV, while PPEH-L blends satisfied the lubricity limits, but not KVlimits. PPEH-L improved low temperature properties of SME biodie-sel whereas PPEH-H had the opposite effect.
Based on these findings, PPEH-H and a mixture of PPEH-H/PPEH-Lare suitable blend components for ULSD in the P10–P50 blend rangeso long as antioxidants are employed.
Acknowledgments
The authors acknowledge Dheeptha Murali (ISTC), Jennifer LDeluhery (ISTC), Benetria N. Banks (USDA-ARS-NCAUR) and Erin L.Walter (USDA-ARS-NCAUR) for excellent technical assistance. Thisstudywas supported in part by seed funding from the Illinois HazardousWaste Research Fund.
References
[1] U.S. Environmental Protection Agency, Office of Solid Waste, Municipal solid wastein the United States, 2009 facts and figures, EPA530-R-10-012, December 2010.
[2] M. Hakkarainen, A.C. Albertsson, Environmental degradation of polyethylene, Ad-vances in Polymer Science 169 (2004) 177–199.
[3] E.A. Howell, S.J. Bograd, C. Morishige, M.P. Seki, J.J. Polovina, On North Pacific circu-lation and associated marine debris concentration, Marine Pollution Bulletin 65(2012) 16–22.
[4] J.G.B. Derraik, The pollution of the marine environment by plastic debris: a review,Marine Pollution Bulletin 44 (2002) 842–852.
[5] D.K.A. Barnes, A. Walters, L. Goncalves, Macroplastics at sea around Antarctica, Ma-rine Environmental Research 70 (2010) 250–252.
[6] B. Lazar, R. Gracan, Ingestion of marine debris by loggerhead sea turtles, Carettacaretta, in the Adriatic Sea, Marine Pollution Bulletin 62 (2011) 43–47.
[7] B. Page, J. McKenzie, R. McIntosh, A. Baylis, A. Morrissey, N. Calvert, et al., Entangle-ment of Australian sea lions and New Zealand fur seals in lost fishing gear and othermarine debris before and after government and industry attempts to reduce theproblem, Marine Pollution Bulletin 49 (2004) 33–42.
[8] C. Muller, K. Townsend, J. Matschullat, Experimental degradation of polymer shop-ping bags (standard and degradable plastic, and biodegradable) in the gastrointes-tinal fluids of sea turtles, Science of the Total Environment 416 (2012) 464–467.
[9] J. Dikgang, A. Leiman, M. Visser, Elasticity of demand, price and time: lessons fromSouth Africa's plastic-bag levy, Applied Economics 44 (2012) 3339–3342.
[11] R.H. Venderbosch, W. Prins, Fast pyrolysis technology development, Biofuels,Bioproducts and Biorefining 4 (2010) 178–208.
[12] S. Yaman, Pyrolysis of biomass to produce fuels and chemical feedstocks, EnergyConversion and Management 45 (2004) 651–671.
[13] D. Mohan, C.U. Pittman Jr., P.H. Steele, Pyrolysis of wood/biomass for bio-oil: a crit-ical review, Energy & Fuels 20 (2006) 848–889.
[14] E. Butler, G. Devlin, D. Meier, K. McDonnell, A review of recent laboratory researchand commercial developments in fast pyrolysis and upgrading, Renewable & Sus-tainable Energy Reviews 15 (2011) 4171–4186.
[15] D.C. Elliot, Historical developments in hydroprocessing bio-oils, Energy & Fuels 21(2007) 1792–1815.
[16] P.M. Mortensen, J.D. Grunwaldt, P.A. Jensen, K.G. Knudsen, A.D. Jensen, A review ofcatalytic upgrading of bio-oil to engine fuels, Applied Catalysis A: General 407(2011) 1–19.
[17] K.D. Maher, D.C. Bressler, Pyrolysis of triglyceride materials for the productionof renewable fuels and chemicals, Bioresource Technology 98 (2007)2351–2368.
[19] A.A. Boeteng, C.A. Mullen, N.M. Goldberg, K.B. Hicks, C.M.McMahan, M.C.Whalen, K.Cornish, Energy-dense liquid fuel intermediates by pyrolysis of guayule (Partheniumargentatum) shrub and bagasse, Fuel 88 (2009) 2207–2215.
[20] C.A. Mullen, A.A. Boeteng, N.M. Goldberg, I.M. Lima, D.A. Laird, K.B. Hicks, Bio-oil andbio-char production from corn cobs and stover by fast pyrolysis, Biomass andBioenergy 34 (2010) 67–74.
[21] A.A. Boeteng, C.A. Mullen, N.M. Goldberg, K.B. Hicks, H.J.G. Jung, J.F.S. Lamb, Produc-tion of bio-oil from alfalfa stems by fluidized-bed fast pyrolysis, Industrial & Engi-neering Chemistry Research 47 (2008) 4115–4122.
[22] A.A. Boateng, C.A.Mullen, N.M. Goldberg, Producing stable pyrolysis liquids from theoil-seed presscakes of mustard family plants: pennycress (Thlaspi arvense L.) andcamelina (Camelina sativa), Energy & Fuels 24 (2010) 6624–6632.
[23] J.P. Cao, X.B. Xiao, S.Y. Zhang, X.Y. Zhao, K. Sato, Y. Ogawa, X.Y. Wei, T. Takarada,Preparation and characterization of bio-oils from internally circulatingfluidized-bed pyrolysis of municipal, livestock, and wood waste, Bioresource Tech-nology 102 (2011) 2009–2015.
[24] S. Ucar, S. Karagoz, A.R. Ozkan, J. Yanik, Evaluation of two different scrap tires as hy-drocarbon source by pyrolysis, Fuel 84 (2005) 1884–1892.
[25] M.N. Islam, M.N. Islam, M.R.A. Beg, The fuel properties of pyrolysis liquid de-rived from urban solid wastes in Bangladesh, Bioresource Technology 92(2004) 181–186.
[27] S.M. Alston, A.D. Clark, J.C. Arnold, B.K. Stein, Environmental impact of pyrolysis ofmixed WEEE plastics part 1: experimental pyrolysis data, Environmental Science& Technology 45 (2011) 9380–9385.
[28] B. Dou, S. Lim, P. Kang, J. Hwang, S. Song, T.U. Yu, K.D. Yoon, Kinetic study in model-ing pyrolysis of refuse plastic fuel, Energy & Fuels 21 (2007) 1442–1447.
[29] F. Murphy, K. McDonnell, E. Bulter, G. Devlin, The evaluation of viscosity anddensity of blends of Cyn-diesel pyrolysis fuel with conventional diesel fuel inrelation to compliance with fuel specifications EN 590:2009, Fuel 91 (2012)112–118.
[30] D.S. Scott, S.R. Czernik, J. Piskorz, D.S.A.G. Radlein, Fast pyrolysis of plastic wastes,Energy & Fuels 4 (1990) 407–411.
[31] H.T. Lin, M.S. Huang, J.W. Luo, L.H. Lin, C.M. Lee, K.L. Ou, Hydrocarbon fuels producedby catalytic pyrolysis of hospital plastic wastes in a fluidized cracking process, FuelProcessing Technology 91 (2010) 1355–1363.
[32] W.L. Yoon, J.S. Park, H. Jung, H.T. Lee, D.K. Lee, Optimization of pyrolytic coprocess-ing of waste plastics and waste motor oil into fuel oils using statistical pentagonalexperimental design, Fuel 78 (1999) 809–813.
[33] Y. Kodera, Y. Ishihara, Novel process for recycling waste plastics to fuel gas using amoving-bed reactor, Energy & Fuels 20 (2006) 155–158.
[34] N. Miskolczi, A. Angyal, L. Bartha, I. Valkai, Fuels by pyrolysis of waste plastics fromagricultural and packaging sectors in a pilot scale reactor, Fuel Processing Technol-ogy 90 (2009) 1032–1040.
[35] A.A. Garforth, Y.H. Lin, P.N. Sharratt, J. Dwyer, Production of hydrocarbons by cata-lytic degradation of high density polyethylene in a laboratory fluidized-bed reactor,Applied Catalysis A: General 169 (1998) 331–342.
[36] G. Elordi, M. Olazar, G. Lopez, M. Amutio, M. Artetxe, R. Aguado, J. Bilbao, Cata-lytic pyrolysis of HDPE in a continuous mode over zeolites catalysts in a conicalspouted bed reactor, Journal of Analytical and Applied Pyrolysis 85 (2009)345–351.
[37] D.R. Vardon, B.K. Sharma, J. Scott, G. Yu, Z. Wang, L. Schideman, Y. Zhang, T.J.Starhmann, Chemical properties of biocrude oil from the hydrothermal liquefactionof Spirulina algae, swine manure, and digested anaerobic sludge, Bioresource Tech-nology 102 (2011) 8295–8303.
[38] D.R. Vardon, B.K. Sharma, G.V. Blazina, K. Rajagopalan, T.J. Strathmann, Thermo-chemical conversion of raw and defatted biomass via hydrothermal liquefactionand slow pyrolysis, Bioresource Technology 109 (2012) 178–187.
[39] D.R. Vardon, B.R. Moser, W. Zheng, K. Witkin, R. Evangelista, T.J. Strathmann,N. Rajagopalan, B.K. Sharma, Complete utilization of spent coffee grounds toproduce biodiesel, ACS Sustainable Chem. Eng. 1 (10) (2013) 1286–1294, http://dx.doi.org/10.1021/sc400145w.
[40] P.A.Z. Suarez, B.R. Moser, B.K. Sharma, S.Z. Erhan, Comparing the lubricity of biofuelsobtained from pyrolysis and alcoholysis of soybean oil and their blends with petro-leum diesel, Fuel 88 (2009) 1143–1147.
[44] K.M. Doll, B.R. Moser, S.Z. Erhan, Surface tension studies of alkyl esters and epoxi-dized alkyl esters relevant to oleochemically based fuel additives, Energy & Fuels21 (2007) 3044–3048.
[45] G. Knothe, A.C. Matheaus, T.W. Ryan III, Cetane numbers of branched andstraight-chain fatty esters determined in a an ignition quality tester, Fuel 82(2003) 971–975.
[46] G. Knothe, K.R. Steidley, Lubricity of components of biodiesel andpetrodiesel. The origin of biodiesel lubricity, Energy & Fuels 19 (2005)1192–1200.
[47] B.R. Moser, Efficacy of specific gravity as a tool for prediction of biodiesel-petroleumdiesel blend ratio, Fuel 99 (2012) 254–261.
[48] C.E. Ejim, B.A. Fleck, A. Amirfazli, Analytical study for atomization of biodiesels andtheir blends in a typical injector: surface tension and viscosity effects, Fuel 86(2007) 1534–1544.
[49] B.R. Moser, Biodiesel production, properties, and feedstocks, In Vitro Cellular & De-velopmental Biology - Plant 45 (2009) 229–266.
[50] J. Van Gerpen, Biodiesel processing and production, Fuel Processing Technology 86(2005) 1097–1107.
