Fuel 150 (2015) 208216

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Desulfurization of pyrolysis fuels obtained from waste: Lube oils, tires

and plastics
Ana-Mara Al-Lal a, David Bolonio a, Alberto Llamas a, Magn Lapuerta b, Laureano Canoira a,
a
Department of Energy & Fuels, ETS Ingenieros de Minas y Energa, Universidad Politcnica de Madrid, Ros Rosas 21, 28003 Madrid, Spain
b
Grupo de Combustibles y Motores, ETS Ingenieros Industriales, Universidad de Castilla La Mancha, Avda. Camilo Jos Cela s/n, 13071 Ciudad Real, Spain

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 Pyrolysis fuels from waste (tires, lube Waste pyrolysis fuels with high sulphur
oils, plastics) have high sulphur
Fraconal disllaon
content.
 Fuels were desulfured by oxidation Light cut Intermediate cut Heavy cut

with peroxiformic acid and methanol Desulfurizaon procedures

extraction.
 Fuels were desulfured with hydrogen Oxidaon Oxidaon
Hydrogen peroxide Hydrogen peroxide
peroxide Fenton catalysts and silica Iron chloride Formic acid
Ultrasound Ultrasound
adsorption. Adsorpon Extracon
Silicagel Methanol
 Fractional distillation followed by
desulfurization could yield heating
diesel. Analycal procedures
XRF spectrometry
GC FPD

a r t i c l e i n f o a b s t r a c t

Article history: An interesting source of hydrocarbon fuels that has attracted considerable attention in the last few years
Received 27 November 2014 is the pyrolysis of waste materials of organic origin, like tires, lube oils and plastics. However, one main
Received in revised form 5 February 2015 drawback, among others, of these pyrolysis fuels is their high sulphur content that makes them inappro-
Accepted 9 February 2015
priate as automotive or heating fuels. This work describes the use of two affordable desulfurization tech-
Available online 18 February 2015
niques without hydrogen to reduce the sulphur content of these three pyrolysis fuels with moderate
success that could make them useful as heating fuels. These desulfurization methods are based on the
Keywords:
oxidation of the sulphur compounds present in these fuels with hydrogen peroxide to more polar sulphur
Waste tires
Waste lube oils
compounds like sulphoxides and sulphones that can be later eliminated by methanol extraction or sil-
Waste plastics icagel adsorption.
Pyrolysis fuels 2015 Elsevier Ltd. All rights reserved.
Desulfurization

1. Introduction gent regulation has forced reneries to greatly improve their

Sulphur compounds remaining in petroleum fractions from top-
ping, hydroskimming or deep conversion processes are a growing
concern for oil reners since in the lapse of a few years the sulphur
specication for engine fuels has dropped from 500 mg/kg to
10 mg/kg in most European countries [1]. This increasingly strin-
Corresponding author. Tel.: +34 91 336 5169; fax: +34 91 336 6948.
E-mail address: laureano.canoira.lopez@upm.es (L. Canoira).
hydrodesulfurization units, increasing the desulfurization rates
and thus consuming huge amounts of hydrogen and energy [2].
In recent years, the search for effective engine fuels from waste
materials has experienced a great push, partly impelled by regula-
tions that try to involve the life cycle of products and materials.
One of the processes which has received more attention is the
pyrolysis of waste materials to produce diesel or heating fuels,
such as waste lube oils, waste tires and waste plastics. However
in all three cases, especially in the rst two ones, the amounts of
sulphur compounds in the fuels obtained from pyrolysis of these

http://dx.doi.org/10.1016/j.fuel.2015.02.034
0016-2361/ 2015 Elsevier Ltd. All rights reserved.
 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216
Nomenclature
GCFPD gas chromatographyame photometric detector
OATS olenic alkylation of thiophenic sulphur
WTPF waste tire pyrolysis fuel
WPPF waste plastic pyrolysis fuel
waste materials is even higher than in the fuels distilled from
crude oil. The reason is that the waste tires contain a great amount
of sulphur coming from the vulcanization process, and the waste
lube oils contain also a great amount of sulphur additives to
improve the lubricity, and this sulphur ends up in the pyrolysis
products as nal compounds. Moreover, the companies carrying
out pyrolysis processes do not usually have the facilities for
hydrodesulfurization like the reneries and also lack of a secure
supply of cheap hydrogen.
There are desulfurization technologies that do not use hydrogen
for catalytic decomposition of organosulphur compounds, and they
are attractive because they attain high levels of sulphur removal.
Among the existing methods the following ones could be
remarked:
(a) Shifting the boiling point by alkylation: after the boiling
point of sulphur-containing compounds has been shifted to
higher values, they can be removed from light fractions by
distillation and concentrated in the low volatile streams of
the renery that could be subjected to hydrodesulfurization.
British Petroleum uses this method for desulfurizing FCC
gasoline streams: the olenic alkylation of thiophenic sul-
phur (OATS) [35].
(b) Desulfurization by extraction: Organosulphur compounds
are more soluble than hydrocarbons in polar solvents. This
method is quite attractive because its applicability requires
low temperature and low pressure, and thus it could be car-
ried out at ambient conditions. Acetone, ethanol, methanol
[6], polyethylene glycol [7] and nitrogen solvents [8] have
shown reasonable extractive desulfurization rates, achieving
between 50% and 90% of sulphur removal.
(c) Desulfurization by conversion and extraction: In these
methods, the hydrocarbon fuel is mixed with an oxidant
(peroxiacetic or peroxiformic acids prepared in situ from
the acids and hydrogen peroxide) at temperatures lower
than 100 C and atmospheric pressure, and the reaction pro-
duct is extracted with a polar solvent like those described in
the precedent paragraph. This combined oxidation-extrac-
tion procedure has yielded excellent results in laboratory
scale experiments, lowering the sulphur content of a straight
run diesel from 4200 ppm to 10 ppm and improving also
other specications like cetane number, density and aromat-
ic content [9]. Some of these combined oxidation-extraction
methods have been proved to commercial or pilot-plant
scale: UniPure process [10] uses a dissolved oxidation cata-
lyst; SulphCo [11,12] applies ultrasound technology for the
oxidation. Photochemical oxidation combined with solvent
extraction has been also used although not at commercial
scale [1316].
(d) Desulfurization by precipitation: The sulphur compounds in
the hydrocarbon streams form with commercial p-acceptors
insoluble charge transfer complexes that could be separated
by ltration, but the efciency of these processes is usually
very low [17,18].
(e) Desulfurization by adsorption: This method is based in the
capacity of a solid adsorbent to selectively adsorb organosul-
phur compounds, either by physical adsorption (adsorptive
209
WLOPF waste lube oil pyrolysis fuel
XRF X-ray uorescence spectroscopy
DS% desulfurization rate
desulfurization) [19,20] or by chemical reaction between
the organosulphur compounds in the fuel and the adsorbent,
usually forming sulphides (reactive adsorption desulfuriza-
tion) [21,22]. In both cases, a critical parameter is the capa-
city of regeneration of the adsorbent.
This work is based in the development of a desulfurization pro-
cess of waste pyrolysis products that could be carried out in small
facilities with a moderate expenditure. The selected desulfuriza-
tion method is the oxidation of the sulphur compounds remaining
in the pyrolysis fraction with hydrogen peroxide, and extraction of
the most polar oxidized sulphur compounds with a polar solvent
like methanol, or topping of the nal fuel by silica adsorption.
These methods, although not new in the literature [2329], allow
lowering the total amount of sulphur and sulphur compounds.
Among the above described methods, two of them have been
selected for desulfurization of the pyrolysis fuels described in this
work: (a) desulfurization by oxidation and methanol extraction
(method c described above) using ultrasounds [3032] and (b)
oxidation with Fenton catalysts, ultrasounds irradiation and
adsorption to purify the fuel product (method e) [3335]. The fac-
tor that has motivated most the selection of the desulfurization
procedures was that, as explained above, these methods could low-
er the high sulphur content of a straight-run gasoil to fall into the
automotive diesel specications, but these expectations were not
fully accomplished because the feedstock used for these experi-
ments was not a straight-run gasoil from a topping crude oil col-
umn but a gasoil cut from pyrolysis processes that present
other added problems like high olen and aromatic contents,
besides its high sulphur content.
2. Experimental
2.1. Materials
Waste tire pyrolysis fuel (WTPF) was supplied by the Instituto
de Carboqumica from the ICB-CSIC in Zaragoza. Waste plastic pyr-
olysis fuel (WPPF) was kindly supplied by the company Elecofasa
in Rivas-Vaciamadrid. Waste lube oil pyrolysis fuel (WLOPF) was
supplied by the Laboratory of Fuels & Petrochemistry from Techno-
getafe Scientic Park.
Hydrogen peroxide of 30%wt and 50% wt used for the experi-
ments was bought from Panreac. Formic acid of 85% wt was pur-
chased from Panreac and methanol grade synthesis was bought
from Sharlau. Copper (II) sulphate, iron (II) sulphate, iron (II) chlo-
ride and iron (III) chloride were purchased from Panreac. Silicagel
60 (Merck) was of 70230 mesh ASTM (0.0630.200 mm) size and
was activated before use. Florisil (0.1500.250 mm) was bought
from Merck. Magnesium silicate Magnesol D-60 was kindly sup-
plied by The Dallas Group of America. Alumina was bought from
Merck.
The ultrasound bath used for the experiments was a model Son-
ic Import ST 840 of 9 L of capacity, 200 W of power and 40 kHz of
frequency. The stirring device was a KPG Teon stirrer powered by
an IKA engine model RW 20D with a maximum stirring speed of
2000 rpm.
210 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216
2.2. Analytical procedures
The analytical method used to follow the decrease in sulphur
has been X-ray uorescence spectroscopy (XRF) (Oxford Lab-
X3000, standard UNE EN ISO 20847) for the total sulphur content
of the samples. As to the variability in the XRF sulphur measure-
ments, these analyses were carried out in a laboratory accredited
by the Spanish National Accreditation Entity (ENAC). The measure-
ment range covers from 0.001% wt. (10 mg/kg) to 5% wt. of sulphur.
The uncertainty of this test varies depending of the detection
range: Ultralow range, 3 mg/kg; low range, 5 mg/kg; medium
range, 0.01%; high range, 0.02%.
It is also interesting to know the molecular structure of the sul-
phur compounds that give this total sulphur amount. For this rea-
son a gas chromatographyame photometric detector GCFPD
(a sulphur selective detector) was also used. Although it was well
known beforehand that an individual identication of these sul-
phur compounds would be almost impossible, at least a range of
complexity of these sulphur compounds could be obtained (stan-
dard ASTM D-5623).
The GCFPD equipment used for this analysis of sulphur com-
pounds was an Agilent 5890 Series II-FPD. The chromatography
column used was a HP-1 semicapillary column (30 m  0.53 mm 
0.25 lm) from Agilent and the carrier gas used was helium at a
column head pressure of 120 kPa. The injector was set at 275 C
and the FPD detector at 300 C. The heating program was: initial
temperature, 60 C (2 min); rate, 15 C/min to 200 C, later 10 C/
min to 300 C and nal time 15 min at 300 C. 1 lL of sample dis-
solved in methanoldichloromethane 1:10 v/v was injected in the
GC in splitless mode.
A light (dimethylsulphone), an intermediate (benzothiophene)
and a heavy (dibenzothiophene) sulphur compounds were chosen
as standards. Probably, these compounds are present in the initial
intermediate cut and in the desulfurized product. These external
GCFPD calibration standards, dimethylsulphone, benzothiophene
and dibenzothiophene were purchased from Aldrich and were dis-
solved in a mixture of methanoldichloromethane 1:10 v/v. The GC
calibration procedure was carried out preparing three external
calibration standards containing 50, 25 and 12.5 mg/L respectively
of each of the calibration compounds mentioned above. This GC
calibration with the condence intervals for it has been included
as Supplementary Material. Fig. 1 shows three chromatograms:
the calibration standard with the three compounds; the intermedi-
ate cut of WLOPF and the light cut from the WLOPF. The heavy cut
could not be injected in the GC because it obstructed the capillary
column and deposited rubbish in the FPD detector. From Fig. 1, it
Fig. 1. GCFPD Chromatogram of standard sulphur compounds represented in
order of increasing retention time: dimethylsulphone, benzothiophene and diben-
zothiophene (blue line). GCFPD Chromatograms of the intermediate (green line)
and light (red line) cuts of WLOPF corresponding to batch 2 (see Table 2) with
3150 mg S/kg and 1490 mg S/kg respectively. (For interpretation of the references
to colour in this gure legend, the reader is referred to the web version of this
article.)
could be inferred that there are lighter sulphur compounds than
dimethylsulphone in the light cut (with retention times <9 min),
but the amounts of sulphur compounds heavier than dibenzothio-
phene (retention times >20 min) in the intermediate cut is moder-
ate although appreciable. Thus, the projected desulfurization
procedures for this intermediate cut seemed very promising. How-
ever, the individual quantication of these compounds by this GC
technique is very complicated and the GC chromatograms were
used solely as a qualitative rapid view of the desulfurization rate
achieved in each case, see Fig. 4 below.
The desulfurization rate (DS%) was based in the analysis of total
sulphur by XRF and expressed as:
Sinitial  Sfinal
DS%  100 1
Sinitial
2.3. Fractional distillation
The crude pyrolysis fuels whose distillation proles after EN ISO
3405 are shown in Table 1 were fractionally distilled at atmospher-
ic pressure (the vacuum distillation was hindered by the formation
great amounts of foams that ooded the distillation column and
contaminated the distillate). Table 2 shows the average sulphur
contents of the crude WLOPF, and the light, intermediate and
heavy cuts obtained from the atmospheric distillation of the crude
pyrolysis product: the heavy cut (boiling point > 300 C) concen-
trated most of the sulphur (around 5000 mg/kg), followed by the
intermediate cut (b.p. 100300 C) that contains between 3000
and 4000 mg/kg and nally the light cut (b.p from room tem-
perature to 100 C) that is the less sulphur contaminated (1500
1800 mg/kg). Fig. 1 shows the GC prole of the sulphur compounds
of the intermediate and light cuts, since the heavy cut could not be
injected in the GC because it obstructed the capillary column and
deposited rubbish in the FPD detector.
Most of the tests described below were carried out on the two
batches of the intermediate fraction of WLOPF. In all the cases,
the results were expressed as desulfurization rates DS (Eq. (1)),
which is a measurement independent of the initial and nal total
sulphur concentration, and the experiments were replicated to
get accurate results. The intermediate cut of WLOPF was selected
as the main fraction to be studied in the desulfurization work
described in this paper, since its boiling range approached that of
fossil diesel.
2.4. Desulfurization procedure: oxidation and methanol extraction
In this rst method, 100 mL of the intermediate cut of the
selected pyrolysis fuel, 14 mL of 85% wt formic acid and 6 mL of
50% wt hydrogen peroxide were mixed in a three necked round
bottomed ask with reux condenser, and heated in the ultra-
sound bath lled with silicone oil at the temperature selected for
the experiment (70, 80 or 90 C), during the specied time, with
additional stirring by a KPG device powered by the mechanical
stirrer at 400 rpm. The ultrasound bath was set at 200 W during
the whole experiment. After the time specied, the crude reaction
product was transferred to a 250 mL decanting funnel and left
decanting for 0.51 h. The lower aqueous layer was removed and
the upper hydrocarbon layer was extracted with 100 mL of metha-
nol with vigorous manual stirring during 2 min. The methanol
extract was discarded and the hydrocarbon layer was extracted
two more times repeating the extraction procedure. Finally, the
hydrocarbon product was dried in rotavapor and analyzed for total
sulphur content by XRF and for sulphur compounds by GCFPD.
The desulfurization of the intermediate cut of the WTPF pyroly-
sis fuel ([S] = 8700 mg/kg) was carried out following this
 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216 211

Table 1
Properties of the waste pyrolysis fuels.

Property Method WTPF WLOPF WPPF Mina Maxa

HHV (MJ/kg) ASTM 24009 42.70 44.96 44.98
LHV (MJ/kg) Estimated from HHV and fuel composition 40.49 42.32 42.35
C (wt%) EN 15104 86.19 84.47 84.68
H (wt%) 10.33 12.85 12.32
O (wt%) 1.86 2.13 2.73
N (wt%) 0.79 0.13 0.23
S (mg/kg) EN 20847 8300 4210 1100 10
FAME (vol%) 0 0 0 7
H/C atomic ratio 1.43 1.81 1.73
Stoichiometric air/fuel ratio (kg/kg) From elemental composition 13.45 14.02 13.84
Molecular weight (kg/kmol) From AspenTech HYSYS 142.5 184.5 211.1
Density at 15 C (kg/m3) EN 12185 917 837 794 820 845
Kinematic viscosity at 40 C (cSt) EN 3104 2.39 2.36 1.50 2.0 4.5
Distillation curve: EN ISO 3405
Initial boiling point (C) 82 110 52
T10 (C) 120 140 125
T20 (C) 146 197 138
T30 (C) 175 255 153
T40 (C) 200 298 167
T50 (C) 233 308 188
T60 (C) 272 321 215
T65 (C) 285 339 228 250
T70 (C) 296 352 241
T80 (C) 305 360 274
T85 (C) 375 296 350
T90 (C) 389 306
T95 (C) 389 360
Final boiling point (C) 321
Saturates (wt%) From TLC for TPL ASTM 6379/04 for diesel 5.5 30.8
Olens (wt%) 44.6
Aromatics (wt%) 65.2 24.6
Polar (wt%) 29.3
CFPP (C) EN 116 9 25 11 10;0
Lubricity (lm WS 1.4) EN 12156 162.6 249.6 188.8 460
Water content (mg/kg) EN 12937 690 225 227 200
TAN (mg KOH/g) ASTM D 664 5
Flash point (C) EN ISO 2719 23.0 25.0 55
Oxidation stability (h) EN 15751 16 0 20
Derived cetane number ASTM D7668 40.9 51
Cetane index EN ISO 4264 17.6 49.0 43.1 46
Total particles (mg/kg) EN 12662 16.2 24
Ash content (wt%) EN ISO 6245 <0.01 0.01
Carbon residue (wt%) EN ISO 10370 0.14 0.30
Colour (u.c.) ASTM D 1500 5.5 2

WTPF = waste tire pyrolysis fuel; WLOPF = waste lube oil pyrolysis fuel; WPPF = waste mixed plastics pyrolysis fuel.
a
Diesel standard EN 590.

general procedure and the failed methanol extraction yielded by

Table 2
Total sulphur concentration in the distillation fractions of WLOPFa. rotavapor distillation of the methanolic solution three fractions:
light (yield 29% wt, total sulphur content 19.2 mg/kg), intermedi-
[S], mg/kg Batch 1 Batch 2
ate (yield 46.6% wt, total sulphur content 440.6 mg/kg, 60% desul-
WLOPF 3905 3800 furization rate) and heavy (yield 23.6% wt, total sulphur content
Heavy cut 5105 4925
1923.6 mg/kg).
Intermediate cut 4125 3150
Light cut 1810 1490
a
Light cut until 100 C, intermediate cut from 100 C to 300 C, heavy cut from
300 C to end point. 2.5. Desulfurization procedure: oxidation and adsorption

In view of the only moderate desulfurization rates obtained by

experimental procedure at 70 C, for 30 min with stirring at
400 rpm and ultrasound irradiation, followed by three consecutive
extractions with methanol (1: 1 v/v WTPF-methanol). The achieved
desulfurization rate was only 53.0%.
The desulfurization of WPPF was carried out on the whole pyr-
olysis fuel without previous distillation step, following exactly this
experimental procedure. However, the methanol extraction failed
because the density of the WPPF (0.794 g/cm3) matches exactly
that of methanol (0.79 g/cm3). Thus, the desulfurized fuel was
separated from the oxidizing reagents by decantation as in the
the experimental procedure 2.4, it was decided to experiment
another desulfurization procedure based in hydrogen peroxide
oxidation catalyzed by Fenton type reagents and solid adsorption
of the oxidized sulphur compounds. The four Fenton type catalysts
tested were copper (II) sulphate, iron (II) sulphate, iron (II) chloride
and iron (III) chloride, and, besides alumina, three other adsorbents
were used for the experiments: silicagel, Florisil and Magnesol
D-60. Ultrasound irradiation was also used in all the experiments,
some of which were carried out in duplicate. The desulfurization
results have a condence interval (95% probability) of 2.5%.
212 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216
The literature references [3335] indicate as the optimal reac-
tion conditions for this case the following: 100 mL of the interme-
diate cut of the selected pyrolysis fuel, 50 mL of 3% wt hydrogen
peroxide and 1 g of solid catalyst were mixed in a 250 mL round
bottomed ask with reux condenser, and the mixture was heated
in the ultrasound bath lled with silicone oil at the desired
temperature (usually 90 C), with stirring at 400 rpm by the KPG
stirring device powered by the IKA engine and ultrasound irra-
diation at a power of 200 W. Afterwards, the crude reaction pro-
duct was ltered in a Bchner funnel to eliminate the solid
catalyst, and the liquid phase was transferred to a decanting funnel
and left decanting for 0.51 h. The lower aqueous layer was dis-
carded and the upper hydrocarbon phase was passed through a
column of the specied adsorbent (6 cm  2.5 cm internal dia-
meter). The eluent product from the adsorption column, without
further purication, was analyzed for total sulphur content by
XRF spectrometry and the sulphur compounds were analyzed by
GCFPD.
3. Feedstocks of the pyrolysis fuels
The pyrolysis fuels chosen to study represent a huge amount of
waste in the world that is, used tires [36], used engine lube oils
[37] and used plastics [38]. The two former, WTPF [39] and WLOPF,
contain a great amount of sulphur, the rst one due to the vulcan-
ization process and the second one due to the sulphur additives of
the lube oils added to improve their lubricity (Table 1). For these
two pyrolysis fuels, the only desulfurization process that could
bring them to the specications of automotive diesel would be
the hydrodesulfurization carried out in reneries at high tem-
perature (320380 C) and high hydrogen pressures (20200 bar).
These strong reaction conditions are difcult to achieve for the
small producers of these pyrolysis fuels, besides the difculty into
nd a cheap and available hydrogen supply. The only possible out-
put for these products would then be to add them to the fueloil
pool, but other properties like the low ash point and the low
initial boiling point in the atmospheric distillation test (among
other failed properties) prevents also this use. For this reason, this
study was started with the aim to bring these fuels to the speci-
cations of the heating gasoil (total sulphur content lower than
1000 mg/kg), improving also other properties, and offering to the
small producers of these fuels a feasible alternative to the
hydrodesulfurization.
The WPPF was a different case since its sulphur content
although still very high (1100 mg/kg) would allow its use as heat-
ing gasoil with a low desulfurization rate. However, this WPPF does
not meet other specications, especially those related to the low
initial boiling point at atmospheric pressure (indicating the pres-
ence of light hydrocarbons) and surprisingly, the nil oxidation sta-
bility (Table 1).
This study was centred in the desulfurization of WLOPF, and the
optimized experimental conditions found for this feedstock were
applied to the other two pyrolysis fuels with high sulphur content,
WTPF and WPPF.
4. Results and discussion
4.1. Initial desulfurization experiments
Among the desulfurization methods found in the literature and
briey described in the introduction section, the peroxiformic acid
oxidation method was chosen for this study since it gave good
results with a straight-run gasoil with sulphur contents similar
to those in this study [9], and the reagents and equipment were
also easily available and affordable for the small producers of pyr-
O
H C
O OH
S
S O
HO
S
O O
Scheme 1. Oxidation of dibenzothiophene by peroxiformic acid or hydroxyl radical.
olysis fuels. As explained in the introduction section, the per-
oxiformic acid oxidizes the sulphur compounds present in the
pyrolysis fuels to more polar compounds like sulphoxides and sul-
phones that are easily extracted with a polar solvent like methanol
or easily adsorbed in a polar adsorbent like silicagel or alumina.
This chemical reaction is briey described in Scheme 1 for diben-
zothiophene. The peroxiformic acid was prepared in situ from for-
mic acid 85% wt and hydrogen peroxide 50% wt, and the crude
WLOPF was reacted without a previous fractional distillation step,
but the desulfurized product was distilled after the double metha-
nol extraction (Table 3). Each experiment was replicated in order
to obtain representative results. The desulfurization rates obtained
were reasonable (73.5%), but much lower than those described in
the literature [9]. Moreover, due to the high amount of hydrogen
peroxide used, all the samples showed water contents higher than
400 mg/kg and also the colour of the samples was out of specica-
tions. Thus, the operation procedure was changed, and the desul-
furization experiments described in Table 4 were carried out on
the intermediate cut of the fractional distillation of crude WLOPF.
Although the desulfurization rates were essentially the same
(69.8%), the water content and the colour of the desulfurized sam-
ples were according to specications, and thus, for the rest of the
experiments, this was the used procedure. The aim of the two
new series of experiments was to optimize the procedures and to
achieve the desulfurization rates described in the literature [30
32].
4.2. Desulfurization by oxidation and methanol extraction
The results of this series of optimization experiments are sum-
marized in Fig. 2. Clearly, an increase of temperature did not favour
the desulfurization, and longer reaction times did not increase sub-
stantially the sulphur elimination, but only the energy cost of the
treatment. In this way, the optimal reactions conditions were:
100 mL WLOPF intermediate cut; 14 mL formic acid 85% wt;
6 mL hydrogen peroxide 50% wt; stirring speed: 400 rpm; extrac-
tion WLOPF-methanol 1:1, 2 min, three times. The DS increased
linearly with the number of liquidliquid methanol extractions
(NE) of the desulfurized product (regression coefcient R2 = 1.00)
(Fig. 3). However, the fourth extraction did not follow this linear
trend and practically did not improve the results of the third one.
The DS did not surpass in any case the 70% yield, far away from
the literature results [3032], probably due to the presence of ole-
nic, aromatic and polar compounds in the WLOPF that consumed
oxidizing reagents and that are absent in straight-run topping cuts.
The application of this desulfurization method to the intermedi-
ate cut of WTPF in the optimal reaction conditions mentioned
above gave only a moderate 53% desulfurization yield, also prob-
ably due to the presence of olenic, aromatic and polar compounds
and the higher sulphur content (8700 mg/kg, practically double
than WLOPF).
 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216 213

Table 3
Results of the desulfurizationa, extraction and distillationb tests.

WLOPF H2O2 50% HCOOH 85% Desulfurization, 1st extraction Desulfurization 2nd extraction % Mean 1st % Mean 2nd
(mL) (mL) (mL) (%) (%) extraction extraction
100 10 3 51.6 61.8 49.6 62.2
100 10 3 47.6 62.6
100 14 3 49.3 62.9 50.7 60.5
100 14 3 52.1 58.1
100 20 6 66.9 75.9 67.4 73.5
100 20 6 67.9 71.1
a
Desulfurization tests were carried out on the whole pyrolysis fuel without previous distillation: [S] initial = 3530 mg/kg. Reactions conditions = 90 C, 4 h, 500 rpm,
extraction (WLOPF-methanol 1:1, 2 min).
b
Distillation was carried out after desulfurization: light cut until 5%v/v distillate, intermediate cut from 5%v/v distillate to 388 C, heavy cut from 388 C to end point.

Table 4
Results of the distillation,a desulfurizationb and extraction tests.

WLOPF (mL) H2O2 50% (mL) HCOOH 85% (mL) Desulfurizationc 1st extraction (%) Desulfurizationc 2nd extraction (%) [H2O] (mg/kg) Colour
100 10 3 46.2 54.5 193 6
100 10 4 44.1 56.6 237 5.5
100 10 5 50.1 58.1 197 6
100 10 6 45.6 59.0 133 6
100 12 4 42.9 56.3 145 4.5
100 14 4 48.6 65.4 192 5
100 12 6 49.5 69.8 160 4
a
Desulfurization tests were carried out on the intermediate fraction of the distillation (light cut until 5%v/v distillate, intermediate cut from 5%v/v distillate to 388 C,
heavy cut from 388 C to end point).
b
[S]initial = 3344 mg/kg. Reactions conditions = 90 C, 4 h, 500 rpm, extraction (WLOPF-methanol 1:1, 2 min).
c
These desulfurization results have a condence interval (95% probability) of 2.5%.

This same desulfurization method was also applied to crude

WPPF with a similar desulfurization yield of 60% but with the
added difculty of the failed methanol extraction, since methanol
and the fuel have the same density. In future experiments, a more
dense polar extraction solvent should be used for this feedstock to
improve the separation of the solvent for recycling and the extract-
ed sulphur compounds [40].

4.3. Desulfurization by oxidation and adsorption

The results of this series of experiments are summarized in

Fig. 2. Reactions conditions for the oxidationmethanol extraction method: 100 mL
WLOPF intermediate cut; 14 mL Formic acid 85% wt; 6 mL Hydrogen peroxide 50%
wt. Stirring speed: 400 rpm. Extraction WLOPF-methanol 1:1, 2 min, three times.
Fig. 3. Extraction WLOPF-methanol 1:1, 2 min.
Table 5. The only results that approached those previously found
in the literature [3335] were with iron (II) and iron (III) chlorides
as catalysts (57.1% and 64.0% desulfurization rates respectively)
and silicagel as adsorbent. The acidication of the reaction medium
with a few drops of sulphuric acid only lowered drastically the
desulfurization yields. The change of the reaction temperature in
10 C above or below 90 C did not modify substantially the yield
nor the increase of the reaction time to 15 or 30 min. The only
parameter that substantially increased the desulfurization yield
was the increase of the height of the silicagel adsorption bed from
6 cm (66.7%) to 9 cm (77.9%) and 12 cm (87.5%), always with an
internal diameter of the adsorption column of 2.5 cm. This para-
meter could have an easy industrial implementation, since silicagel
adsorption columns could be installed by triplicate, one in the
adsorption mode to eliminate the oxidized sulphur compounds,
the second in the regeneration mode burning the adsorbed sulphur
compounds in an air stream at high temperature, and the third in
the conditioning mode preparing it from the high temperature of
the regeneration to the low temperature of the adsorption, like
in the pressure swing adsorption method used for hydrogen puri-
cation in reneries [41]. The effect of the ultrasound irradiation is
clear, since its absence lowered the desulfurization yield to 52.2%,
but the methanol extraction prior to adsorption in this method has
a small detrimental effect and the use of silicagel without previous
activation lowered the yield but in an insignicant amount
214 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216

Table 5
Results with the Fenton type catalysts and the adsorption method.

Temperature (C) Time (min) Catalyst Adsorbent Adsorbent bed height (cm) % Desulfurizationa
90 8 FeSO4 Silicagel 6 40.6
Florisil 31.6
Alumina 34.9
Magnesol D-60 33.8
90 8 CuSO4 Silicagel 6 39.4
Florisil 24.1
Alumina 29.5
Magnesol D-60 23.4
90 8 FeCl2 Silicagel 6 57.1
Florisil 49.1
Alumina 55.8
Magnesol D-60 35.9
90 8 FeCl3 Silicagel 6 64.0
Florisil 33.1
Alumina 34.4
Magnesol D-60 27.3
90 8 FeCl3/H2SO4 Silicagel 6 33.0
Florisil 17.3
Alumina 21.3
Magnesol D-60 15.4
80 8 FeCl3 Silicagel 6 65.7
90 8 FeCl3 Silicagel 6 66.7
100 8 FeCl3 Silicagel 6 60.0
90 8 FeCl3 Silicagel 9 77.9
90 8 FeCl3 Silicagel 12 87.5
90 8 FeCl3 Silicagel 6 52.2b
90 8 FeCl3 Silicagel 6 62.9c
90 8 FeCl3 Silicagel 6 64.0d
a
These desulfurization results have a condence interval (95% probability) of 2.5%.
b
Without ultrasound irradiation.
c
With methanol extraction before adsorption.
d
Silicagel without previous activation.

Fig. 5. GC-FPD Chromatograms of sulphur compounds of a pyrolysis fuel from

waste lube oil before (blue line) and after (red line) the desulfurization process. (For
interpretation of the references to colour in this gure legend, the reader is referred
Fig. 4. GCFPD Chromatograms of poor desulfurization results (48.1%, blue line),
to the web version of this article.)
moderate desulfurization results (66.7%, red line) and good desulfurization results
(77.3%, green line). (For interpretation of the references to colour in this gure
legend, the reader is referred to the web version of this article.)

to the sulphur specication for heating fuel (sulphur less than

(Table 5). Fig. 4 shows the chromatograms of three samples result-
ing from this method, selected as examples of good, moderate and
poor desulfurization rates.
Fig. 5 shows the selective chromatograms of sulphur com-
pounds of a pyrolisys fuel from waste lube oil before and after
the desulfurization process carried out in this work by the oxida-
tionadsorption method, the total amount of sulphur measured
by XRF being indicated in the insert. The typical desulfurization
rate for a diesel fuel range fraction is around 74.5%, reducing the
amount of sulphur from around 4125 mg/kg to 1050 mg/kg, closer
1000 mg/kg), but very far from the hydrodesulfurization rates of
more than 99% found in oil reneries.
The aim of the work was to bring these fuels closer to the
specications of the heating gasoil. To achieve this objective, some
of the properties have been improved like the ash point, the
cetane index, the atmospheric distillation curve, the HHV and
LHV, the colour, the carbon residue, the oxidation stability and
the water content, but as a consequence of the distillation process
to eliminate the lighter compounds, some other properties like the
density, the viscosity and mainly the cold ow properties like the
cloud point and the CFPP have worsened. Table 6 shows a compar-
 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216 215

Table 6
Properties of the WLOPF before and after desulfurization.

Property Method Before After Da Minb Maxb

HHV (MJ/kg) ASTM 24009 44.96 46.00 +1.04
LHV (MJ/kg) Estimated from HHV and fuel composition 42.32 43.33 +1.01
C (wt%) EN 15104 84.47 85.60 +1.13
H (wt%) 12.85 12.61 0.24
O (wt%) 2.13 1.59 0.54
N (wt%) 0.13 0.10 0.03
S (mg/kg) EN 20847 4210 998 3212 1000
FAME (vol%) 0 0
H/C atomic ratio 1.81 1.76 0.05
Stoichiometric air/fuel ratio (kg/kg) From elemental composition 14.02 14.09 +0.07
Molecular weight (kg/kmol) From AspenTech HYSYS 184.5 220.8 +36.3
Density at 15 C (kg/m3) EN 12185 837 845 +8 900
Kinematic viscosity at 40 C (cSt) EN 3104 2.36 3.10 +0.74 7.00
Distillation curve: EN ISO 3405
Initial boiling point (C) 110 160 +50
T10 (C) 140 208 +68
T20 (C) 197 235 +38
T30 (C) 255 260 +5
T40 (C) 298 282  16
T50 (C) 308 302 6
T60 (C) 321 320 1
T65 (C) 339 338 1 250
T70 (C) 352 354 +2
T80 (C) 360 362 +2 390
T85 (C) 375 363 12
T90 (C) 389 372 17
T95 (C) 389 385 4
Final boiling point (C) 386 Note
Saturates (wt%) From TLC for TPL ASTM 6379/04 for diesel 30.8 57.3 +26.2
Olens (wt%) 44.6 40.4 4.2
Aromatics (wt%) 24.6 2.3 22.3
Cloud point, C ASTM D2500 24 13 +11 4
CFPP (C) EN 116 25 15 +10
Lubricity (lm WS 1.4) EN 12156 249.6 369.8 +120.2
Water content (mg/kg) EN 12937 225 176 49
Water and sediments,% vol EN 51083 <0.01 <0.01 0.10
Flash point (C) EN ISO 2719 25 65 +40 60
Oxidation stability (g/m3) ASTM D 2274 57 21 36
Copper strip corrosin, class ASTM D130 1a 1a Class2
Cetane index EN ISO 4264 49.0 59.0 +10.0
Total particles (mg/kg) EN 12662 16.2 13.5 2.7
Ash content (wt%) EN ISO 6245 <0.01 <0.01
Carbon residue (wt%) EN ISO 10370 0.14 <0.01 0.13 0.35
Colour (u.c.) ASTM D 1500 5.5 1.5 4.0 2.0

WLOPF = waste lube oil pyrolysis fuel.

a
D = difference in the value of this property after desulfurization before desulfurization.
b
Specications for heating diesel fuel in Spain (RD 1088/2010).

ison of some of these properties for WLOPF. It is especially notice-
able the worsening of the lubricity of the fuel as a consequence of
the elimination of the sulphur aromatic compounds.
5. Conclusions
for automotive diesel and in some cases also for heating diesel.
The procedure described in this work could allow the pyrolysis
fuels from waste materials to be blended with another waste origin
material like animal fat biodiesel (between 35 and 49 mg/kg of sul-
phur) and to be sold as heating fuel.

Three pyrolysis fuels from different waste materials have been

characterized and studied as possible fuels or fuel components
for heating and automotion. Among other parameters that hin-
dered their use because they are out of specications, the main
one is the high sulphur content of most of them. Two different
desulfurization procedures have been tested, based both in oxida-
tion of the sulphur compounds followed either by methanol
extraction of the more polar oxidized sulphur compounds in one
of the procedures and silicagel column adsorption of the more
polar oxidized sulphur compounds in the other. Both procedures
gave moderate to good yields of desulfurization, but very far away
from the hydrodesulfurization rates achieved in reneries, and
they leave still these pyrolysis fuels out of sulphur specications
Acknowledgements
The authors wish to thank: Ramn Murillo from the Instituto de
Carboqumica of ICB-CSIC in Zaragoza for the supply of WTPF, Ele-
cofasa in Rivas-Vaciamadrid for the supply of WPPF, the Laboratory
of Fuels & Petrochemistry of The Gmez-Pardo Foundation for the
supply of WLOPF, Emilio Garca, Yolanda Cereceda, Natalia Timn
and Irene Gutierrez for their technical help in the development
of the experimental tests, La Caixa Foundation for a research grant
to David Bolonio and the Spanish Ministry of Economy and Com-
petitiveness for his nancial support through project WOLF (Ref.
ENE2013-48602-C3-1-R).
216 A.-M. Al-Lal et al. / Fuel 150 (2015) 208216
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.fuel.2015.02.034.
References
[1] Royal Decree 1088/2010, BOE 4th September 2010, 215: 7643640.
[2] Lluch Urpi J. Tecnologa y margen de reno del petrleo, 1st ed. Ediciones Daz
de Santos SA, Madrid; 2008.
[3] Wiltshire J. BP Technol Mag 2000;32:10.
[4] Burnett PT, Huff GA, Pradhan VR, Hodges M, Glassett JA, McDaniel SG, Hurst P.
The European rening technology conference, Rome, Italy; November 1315;
2000.
[5] BP tests alkylation/fractionation process for sulphur removal. Octane Week
9.10.2000.
[6] Funakoshi I, Aida T. US Patent 5,753,102.
[7] Forte P. US Patent 5,582,714.
[8] Horii Y, Onuki H, Doi S, Mori T, Takatori T, Sato H, Ookuro T, Sugawara T. US
Patent 5,494,572.
[9] Bonde SE, Chapados D, Gore WL, Dolbear G, Skov E. NPRA 2000 Annual
Meeting, AM-00-25, San Antonio, Texas; March 2628; 2000.
[10] Levy RE, Rappas AS, DeCanio SJ, Nero VP. NPRA 2001 Annual Meeting, AM-01-
10, New Orleans; March 1820; 2001.
[11] Peckham J. Octane Week 2001; January 5th.
[12] Shake the sulphur right out of fuel. Chem Eng Prog; 2001, May 1st.
[13] Shiraishi Y, Hirai T, Komasawa I. A deep desulfurization process for light oil by
photochemical reaction in an organic two-phase liquidliquid extraction
system. Ind Eng Chem Res 1998;37:20311.
[14] Shiraishi Y, Taki Y, Hirai T, Komasawa I. Visible light-induced deep
desulfurization process for light oils by photochemical electron-transfer
oxidation in an organic two-phase extraction system. Ind Eng Chem Res
1999;38:33108.
[15] Shiraishi Y, Taki Y, Hirai T, Komasawa I. Visible light-induced desulfurization
process for catalytic-cracked gasoline using an organic two-phase extraction
system. Ind Eng Chem Res 1999;38:453844.
[16] Shiraishi Y, Hirai T, Komasawa I. Photochemical desulfurization and
denitrogenation process for vacuum gas oil using an organic two-phase
extraction system. Ind Eng Chem Res 2001;40:293303.
[17] Meille V, Shulz E, Vrinat M, Lemaire M. A new route towards deep
desulfurization: selective charge transfer complex formation. Chem Commun
1998;3:3056.
[18] Milenkovic A, Shulz E, Meille V, Lofferda D, Forissier M, Vrinat M, et al.
Selective elimination of alkyldibenzothiophenes from gas oil by formation of
insoluble charge-transfer complexes. Energy Fuels 1999;13:8817.
[19] Salem ABSH. Naphtha desulfurization by adsorption. Ind Eng Chem Res
1994;33:33640.
[20] Salem ABSH, Hamid HS. Removal of sulphur compounds from naphta solutions
using solid adsorbents. Chem Eng Technol 1997;20:342.
[21] Greenwood GJ, Kidd D, Reed L. NPRA 2000 Annual Meeting, AM-00-12, San-
Antonio, Texas; March 2628; 2000.
[22] Parkinson G. Reneries crack down on sulphur. Chem Eng 2000;8:45.
[23] Babich IV, Moulijn JA. Science and technology of novel processes for deep
desulfurization of oil renery streams: a review. Fuel 2003;82:60731.
[24] Farhat Ali M, Al-Malki A, El-Ali B, Martinie G, Siddiqui MN. Deep
desulphurization of gasoline and diesel fuels using non-hydrogen consuming
techniques. Fuel 2006;85:135463.
[25] Farhat Ali M, Al-Malki A, Ahmed S. Chemical desulfurization of petroleum
fractions for ultra-low sulfur fuels. Fuel Process Technol 2009;90:53644.
[26] Dehkordi AM, Sobati MA, Ali Nazem M. Oxidative desulfurization of non-
hydrotreated kerosene using hydrogen peroxide and acetic acid. Chin J Chem
Eng 2009;17:86974.
[27] Shakirullah M, Ahmad W, Ahmad I, Ishaq M. Oxidative desulphurization study
of gasoline and kerosene: role of some organic and inorganic oxidants. Fuel
Process Technol 2010;91:173641.
[28] Muzica M, Sertic-Biondaa K, Gomzia Z, Podolskib S, Telenb S. Study of diesel
fuel desulfurization by adsorption. Chem Eng Res Des 2010;88:48795.
[29] Selvavathi V, Chidambaram V, Meenakshisundaram A, Sairam B, Sivasankar B.
Adsorptive desulfurization of diesel on activated carbon and nickel supported
systems. Catal Today 2009;141:99102.
[30] Mei H, Meil BW, Yen TF. A new method for obtaining ultra-low sulphur diesel
fuel via ultrasound assisted oxidative desulfurization. Fuel 2003;82:40514.
[31] Chen TC, Shen YH, Lee WJ, Lin CC, Wan MW. The study of ultrasound-assisted
oxidative desulfurization process applied to the utilization of pyrolysis oil
from waste tires. J Clean Prod 2010;18:18508.
[32] Duarte FA, Mello PA, Bizzi CA, Nunes MAG, Moreira EM, Alencar MS, et al.
Sulfur removal from hydrotreated petroleum fractions using ultrasound-
assisted oxidative desulfurization process. Fuel 2011;90:215864.
[33] Flores R, Rodas A, Chavarria W. Desulfurization of fuel oils using an advanced
oxidation method. Prepr Pap Am Chem Soc Div Fuel Chem 2004;49:341.
[34] Dai Y, Qi Y, Zhao D, Zhang H. An oxidative desulfurization method using
ultrasound/Fentons reagent for obtaining low and/or ultra-low sulfur diesel
fuel. Fuel Process Technol 2008;89:92732.
[35] Bhasarkar JB, Chakma S, Moholkar VS. Mechanistic features of oxidative
desulfurization using Sono-FentonPeracetic acid (ultrasound/Fe2+ 
CH3COOHH2O2) system. Ind Eng Chem Res 2013;52:903847.
[36] Lpez G, Olazar M, Aguado R, Bilbao J. Continuous pyrolysis of waste tires in a
conical spouted bed reactor. Fuel 2010;89(8):194652.
[37] http://www.epa.gov/osw/conserve/materials/usedoil/.
[38] Al-Salem SM, Lettieri P, Baeyens J. Recycling and recovery routes of plastic
solid waste (PSW): a review. Waste manage 2009;29(10):262543.
[39] Martnez JD, Lapuerta M, Garca-Contreras R, Murillo R, Garca T. Fuel
properties of tire pyrolysis liquid and its blends with diesel fuel. Energy
Fuels 2013;27:3296305.
[40] Rodrguez-Cabo B, Rodrguez H, Rodil E, Arce A, Soto A. Extractive and
oxidative extractive desulfurization of fuels with ionic liquids. Fuel
2014;117:8829.
[41] Meyers RA. Handbook of Petroleum Rening Processes. 3rd ed. New
York: McGraw-Hill; 2004.