A novel oxidative desulfurization process to remove refractory sulfur compounds from diesel fuel

A novel oxidative desulfurization process to remove refractory

sulfur compounds from diesel fuel

Jeyagowry T. Sampanthar *, Huang Xiao, Jian Dou, Teo Yin Nah,Xu Rong, Wong Pui Kwan

Applied Catalysis Technology Group, Institute of Chemical and Engineering Sciences, Agency for Science,

Technology and Research (A*STAR), No. 1, Pesek Road, Jurong Island, Singapore 627833, Singapore

Received 17 June 2005; received in revised form 12 September 2005; accepted 12 September 2005

Available online 25 October 2005

Abstract

Manganese and cobalt oxide catalysts supported on g-Al2O3 have been found to be effective in catalyzing air oxidation of the sulfur impurities in

diesel to corresponding sulfones at a temperature range of 130–200 8C and atmospheric pressure. The sulfones were removed by extraction with polar

solvent to reduce the sulfur level in diesel to as low as 40–60 ppm. Oxidation of model compounds showed that the most refractory sulfur compounds in

hydrodesulfurization of diesel were more reactive in oxidation. The oxidative reactivity of model impurities in diesel follows the order: trialkyl-

substituted dibenzothiophene > dialkyl-substituted dibenzothiophene > monoalkyl-substituted dibenzothiophene > dibenzothiophene.

# 2005 Elsevier B.V. All rights reserved.

Keywords: Oxidative desulfurization; Diesel; Sulfur; Catalyst; MnO2/g-Al2O3; Co3O4/g-Al2O3; Solvent extraction

www.elsevier.com/locate/apcatb

Applied Catalysis B: Environmental 63 (2006) 85–93

1. Introduction

Deep desulfurization of diesel fuel has become an important

research subject due to the upcoming legislative regulations to

reduce sulfur content in most western countries. The US Clean

Air Act Amendments of 1990 and the new regulations by the

US Environmental Protection Agency (EPA) and government

regulations in many countries call for the production and use of

more environment-friendly transportation fuels with lower

contents of sulfur and aromatics. The demand for transportation

fuels has been increasing in most countries for past two

decades. For example, US Environmental Protection Agency

has set up guidelines to limit the sulfur content of diesel fuel to

15 ppm by 2006 [1]. Conventional hydrodesulfurization (HDS)

process has been employed by refineries to remove organic

sulfur from fuels for several decades and the lowest sulfur

content achieved by such process in the fuels is around

500 ppm. However, to meet the challenges of producing ultra-

clean fuels, especially with sulfur content lower than 15 ppm,

both capital investment and operational costs would be rather

high due to more severe operating conditions [2]. Consequently,

* Corresponding author. Tel.: +65 67963819; fax: +65 63166182.

E-mail address: [email protected] (J.T. Sampanthar).

0926-3373/$ – see front matter # 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.apcatb.2005.09.007

several alternative approaches have been used, such as bio-

desulfurization [3], selective adsorption [4], extraction by

ionic-liquid [5] and oxidative desulfurization (ODS) [6–14].

Various studies on the ODS process have reported the use of

differing oxidizing agents and catalysts, such as H2O2/acetic

acid [7] and H2O2/formic acid [8], H2O2/heteropolyacids [9],

H2O2/inorganic solid acids [10], NO2/heterogeneous catalysts

[11], ozone/heterogeneous catalysts [12], tert-butylperoxides/

heterogeneous catalysts [13] and O2/aldehyde/cobalt catalysts

[14]. The ODS process is usually carried out under mild

conditions which present competitiveness over the conven-

tional HDS process [15]. In this process, the sulfur compounds

present in diesel are oxidized by the oxidizing agent to give rise

to the corresponding sulfones. These sulfones are highly

polarized compounds, such that they are removed from the

diesel by subsequent solvent extraction using water-soluble

polar solvents, such as NMP, DMF, DMSO and MeOH, etc.

[15]. By combination of the processes, the sulfur content of the

diesel can be reduced to �50 ppm [16]. Scheme 1 shows the

oxidation of organic sulfur compounds. The resulting sulfones

can be removed by either extraction and/or adsorption.

Here, we report the effective use of air as an environmentally

benign and low-cost oxidant to oxidize the sulfur compounds in

diesel at ambient pressure and moderate temperature in the

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–9386

Scheme 1.

presence of heterogeneous based simple transition metal oxides

loaded on g-Al2O3.

2. Experimental

2.1. Materials

Untreated diesel with a sulfur content in the range of 430–

465 ppm was obtained from Shell Petroleum Corporation,

Singapore. Co(NO3)�6H2O, Mn(CH3COO)2�4H2O, diben-

zothiophene (DBT), 4-methyl-dibenzothiophene (4-MDBT),

4,6-dimethyl-dibenzothiophene (4,6-DMDBT), 4,6-diethyl-

dibenzothiophene (4,6-DEDBT) and n-tetradecane were purc-

hased from Sigma–Aldrich, Singapore and used without further

purification. Model diesel was prepared by adding equimolar

amounts of dibenzothiophene, 4-methyl-dibenzothiophene,

4,6-dimethyl-dibenzothiophene and 4,6-diethyl-dibenzothio-

phene to n-tetradecane to make up a solution with a total sulfur

content of 400 ppm.

2.2. Catalyst preparation

A 10 g of g-Al2O3 pellet (obtained from Singapore Catalyst

Technology Center, diameter �3–4 mm, length �6–10 mm

with a specific surface area and a total pore volume of

370 m2 g�1 and 0.87 ml g�1, respectively) was impregnated

with cobalt nitrate and/or manganese acetate aqueous solutions

by an incipient wetness method. The total metal oxide loading

with respect to g-Al2O3 ranged from 2 to 13 wt%. The

impregnated samples were left on a roller which was set at

25 rpm for 18 h to obtain better dispersion. The water content of

the samples were removed and dried at 120 8C in the oven for

18 h followed by calcination in a static furnace at 550 8C for 5 h

with a ramp of 5 8C min�1. Other catalysts, such as W, Ni, Fe

and Cu, were also prepared in a similar manner.

2.3. Catalyst characterization

The physical and chemical properties of the prepared

catalysts were characterized by various analytical techniques.

The monolayer deposition of metal oxides on g-Al2O3 was

confirmed using powder X-ray diffraction (XRD) technique

using a Bruker AXS D8 Advance instrument with Cu Ka

radiation at 40 kV and 40 mA. The N2 adsorption–desorption

isotherm of the catalysts were studied using an Autosorb-1 at

77 K. Prior to the measurement, the calcined catalysts were

degassed at 300 8C. The thermal behavior of uncalcined

catalysts were studied by thermo gravimetric analysis (TGA)

using Universal V2.5H TA, Model SDT 2690 instrument under

laminar flow of air with a flow rate of 90 ml min�1. X-ray

photoelectron spectroscopy (XPS) investigation was conducted

on a VGESCALAB 250 spectrometer using monochromatic Al

Ka, X-ray source (1486.6 eV) at a constant analyzer pass

energy of 20.0 eV. All binding energies were referenced to the

C 1s peak arising from adventitious carbon (BE 285.0 eV). The

metal oxides composition and loading in the catalysts were also

characterized and confirmed by ICP (Model Vista-MPX) and

SEM-EDX (Model Jeol JSM-6700F).

2.4. Analysis of S—content in the model and real diesel

Sulfur content of the model and real diesel samples were

analyzed by XRF and GC-AED. The total sulfur content in the

model and real diesel samples were analyzed using XRF

(Bruker AXS S4 Exporer), which was calibrated with six liquid

calibration standards (obtained from AccuStandard), and

concentration ranging from 0 to 500 ppm sulfur by wt%.

The 10 ml of samples were placed into 40 mm diameter plastic

cells equipped with 2.5 mm MYLAR polyester film window.

Each cell was vented to prevent the polyester film window from

bulging during the analysis. The samples were then placed in

the automatic sample chamber and the optical path of the XRF

was flushed with helium gas prior to the measurement.

A 6890 GC coupled with atomic emission detector (JAS-

AED) was used to identify the various sulfur compounds and

their concentration. The GC was equipped with a split/splitless

injection port and operated in split mode. A 30 m � 0.32 mm

i.d. � 1 mm film thickness HP-1 MS capillary column was used

for separation as it has a lower specified rate of column bleed

than conventional methyl silicone capillary columns. Hydrogen

and oxygen gases were used as reagent gases for both carbon

(179 nm) and sulfur (181 nm). In order to improve the sulfur

selectivity over carbon, the AED gas flows (hydrogen, oxygen

and helium makeup) were optimized to minimize interferences

from hydrocarbons. The samples volume of 1 ml was injected

without any solvent dilution. The GC-AED instrument was

calibrated with sulfur in diesel fuel SRM 2724 obtained from

National Institute of Standards and Technology (NIST),

Reference Material Department, US. The SRM 2724 is a

commercial no. 2-distillate fuel oil as defined by ASTM with

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–93 87

Table 1

BET surface areas and total pore volume of the prepared catalysts after

calcinations at 550 8C

Catalysta Surface area (m2 g�1) TPV (ml)

g-Al2O3 377 0.87

�2%MnO2/g-Al2O3 361 0.86

�5%MnO2/g-Al2O3 350 0.86

�8%MnO2/g-Al2O3 331 0.78

�11%MnO2/g-Al2O3 317 0.80

�13%MnO2/g-Al2O3 305 0.77

�2%Co3O4//g-Al2O3 368 0.79

�5%Co3O4//g-Al2O3 350 0.76

�8%Co3O4//g-Al2O3 323 0.76

�5%MnO2/3%Co3O4//g-Al2O3 322 0.72

�3%MnO2/3%Co3O4//g-Al2O3 331 0.75

�3%MnO2/5%Co3O4//g-Al2O3 310 0.72

a All the above samples were calcined at 550 8C under static air and degas at

300 8C for 5 h before measurements.

the certified total sulfur content of 425 ppm. In addition to the

SRM 2724, the AccuStandard (also obtained from NIST) with

sulfur content in diesel fuel 0, 100, 200, 300, 400 and 500 ppm

samples were also used for the calibration.

2.5. Solvent extraction on real diesel without oxidative

treatment

Solvent extraction studies for the removal of sulfur

compounds in untreated diesel (obtained from Shell Petroleum

Corporation, Singapore with a sulfur content in the range of 430–

465 ppm) were carried out with four different organic solvents of

different polarities, such as acetonitrile (AcN), dimethylfor-

amide (DMF), 1-methyl-2-pyrrolidinone (NMP) and methonal

(MeOH). A 25.0 ml of untreated diesel were mixed with the

known volume of polar organic solvents to determine the

efficiency of solvent extraction. The diesel–solvent mixture was

stirred for 30 min before separating the two layers. After

extraction by the respective polar solvents, the sulfur content in

the diesel was measured by XRF and GC-AED.

2.6. Catalytic oxidation followed by solvent extraction on

model diesel

The oxidation experiments in this study were carried out with

20.0 ml of model diesel in a refluxed round bottom flask.

Approximately, 20–30 mg of g-Al2O3 supported Mn and/or Co

oxides in the form of pellets were used as catalysts. The reactions

were carried out at a temperature range of 90–180 8C, during

which air was introduced via a gas disperser at a constant flow

rate of 100 ml min�1 while the reaction mixture was stirred

throughout the experiment. Awater-cooled reflux condenser was

mounted on top of the reaction flask to prevent solvent loss and

Fig. 1. X-ray photoelectron spectroscopy analysis of metals oxides loaded on gamma

�3%MnO2/�5%Co3O4/g-Al2O3; (D) �8%Co3O4/g-Al2O3.

serve as an air outlet. The progress of the reaction was monitored

periodically withdrawing 0.5 ml aliquots of the reaction mixture

for GC-AED analysis. Blank experiments were carried out with

model diesel and pure support of g-Al2O3 under exactly similar

experimental conditions.

The oxidized products in the model diesel were extracted

with NMP or methanol after the completion of the reaction.

The reacted model diesel was mixed with one of these polar

solvents at different volume ratio (e.g. diesel:polar solvent = 4:1

when NMP as a solvent, 1:1 when MeOH as solvent) and was

magnetically stirred for 30 min. The mixture was then transferred

into a separating funnel and separated into two layers of model

diesel and polar solvent which were analyzed for sulfur

compounds using GC-AED. When MeOH was used as solvent,

it was removed from solvent–sulfone mixture using rotary

-alumina: (A) �11%MnO2/g-Al2O3; (B) �5%MnO2/�3%Co3O4/g-Al2O3; (C)

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–9388

Fig. 2. Sulfur-specific gas chromatograms of model diesel (air oxidation of model diesel (400 ppm sulfur) catalyzed MnO2 (11%) loaded on g-alumina support;

solvent extraction was carried out using NMP:oxidized diesel = 1:1, single extraction).

evaporator and the product of sulfones mixture was precipitated

at the bottom of the flask.

2.7. Catalytic oxidation by Mn and/or Co oxides supported

on g-Al2O3 followed by solvent extraction on real diesel

A 150.0 ml real diesel underwent oxidative desulfurization

reaction in the presence of about 100 mg of catalyst at

temperature range of 130–200 8C in a two-necked round

bottom flask. The reaction mixture was magnetically stirred to

ensure a good mixing and bubbled with purified air at constant

flow of 100 ml min�1. The reaction mixture was periodically

Fig. 3. Conversion % of the thiophenic compounds to the corresponding sulfone

in model diesel (catalyst: 11%MnO2 loaded in g-Al2O3, temperature 150 8C).

sampled and analyzed using GC-AED and reaction was ceased

after about 18 h. The oxidized diesel was then cooled to room

temperature and 25.0 ml of reacted diesel was treated with

varying volume of different solvents for solvent extraction.

Sulfur content of the extracted oxidized real diesel was

measured by XRF and GC-AED. Similar reaction and solvent

extraction method were carried out with different loading of

both Mn and/or Co oxide catalysts.

3. Results and discussion

3.1. Characterization of catalyst

The TGA studies showed that the most of the metal salts

loaded on the g-Al2O3 converted into their corresponding

oxides at below 500 8C under laminar flow of air. Table 1

summarizes the specific surface area and total pore volume of

the prepared catalysts. It shows that the specific surface area in

the series considerably lowered from 377 m2 g�1 for pure

g-Al2O3 support to the lowest value of 305 m2 g�1 for the

sample loaded with the maximum amount of transition metal

oxide (�13%MnO2/g-Al2O3). The total pore volume of the

calcined samples also decreases as the loading of the transition

metal oxides increases. The decreasing behavior of both surface

area and total pore volume with the increasing loading of the

metal oxides are consistent due to the possible blockage of the

inner pores, especially the smaller ones, and dilution of the

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–93 89

Fig. 4. Sulfur-specific gas chromatograms of real diesel (air oxidation of real diesel diesel (�450 ppm sulfur) catalyzed MnO2 (11%) loaded on g-alumina support;

solvent extraction was carried out using NMP:oxidized diesel = 1:1, single extraction).

Fig. 5. Conversion % thiophenic compounds to corresponding sulfone in real

diesel (catalyst: 11%MnO2 loaded in g-Al2O3, temperature 150 8C).

initial support material, g-Al2O3, by the uniformly dispersed

and dense metal oxide, MnO2 and Co3O4, phase.

The absence of characteristic diffraction peaks in XRD

patterns confirmed that the deposition of metal oxides on the

support g-Al2O3 were in the form of amorphous layer. The data

obtained for the actual loadings of metal oxides from ICP

analysis, XRF and SEM coupled with EDX were almost equal

to the initial calculated values. This confirmed the calcinations

process and its process conditions were optimum and virtually

all the metal oxides coated on the support material.

XPS spectra with binding energies (eV) for metal elements are

shown in Fig. 1. The binding energies of Mn 2p (641.5 eV) and

Co 2p (780.4 eV) for manganese oxide sample A and cobalt

oxide sample D are consistent with the formation of MnO2 and

Co3O4 in these samples. However, for samples B and C which

contains binary Mn–Co oxides coatings, there is a pronounced

shift from +0.7 to +1.2 eV for Mn 2p which could be attributed to

the interaction of manganese with alumina support. The positive

shift of Al 2p binding energies in sample B and C also suggests

that existence of interaction between the coated metal oxides and

the g-Al2O3 support when both Mn and Co oxides are present.

3.2. Selective catalytic sulfur oxidation followed by solvent

extraction on model diesel

As shown by the sulfur-specific gas chromatograms in Fig. 2

and % of conversion versus time in Fig. 3, the thiophenes

conversion increased with time and it reached its maximum

conversion of �80–90% at 8 h. Fig. 3 also shows that the

oxidative reactivity of the model thiophene compounds follows

the order of 4,6-dEDBT > 4,6-dMDBT > 4-MDBT > DBT.

The observed order of reactivity is opposite to that observed in

the hydrodesulfurization process where the most sterically

hindered thiophenes, 4,6-dEDBT and 4,6-dMDBT, are the least

reactive. Apparently, the increased electron density of the sulfur

atoms in disubstituted thiophenes can overcompensate for the

steric hindrance of the C4 and C6 alkyl groups in the oxidative

process.

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–9390

Table 2

Sulfur content analysis results after solvent extraction of diesel with and without

oxidation

Catalysta Extraction

solvent (vol)

S content (ppm)b

(treated diesel)

Diesel, no oxidation No extraction 430

Diesel, no oxidation AcN (10 ml) 310

Diesel, no oxidation DMF (10 ml) 226

Diesel, no oxidation NMP (10 ml) 219

Diesel, no oxidation MeOH (25 ml) 314

�2%Co3O4/g-Al2O3 AcN (10 ml) 237

�2%Co3O4/g-Al2O3 DMF (10 ml) 146

�2%Co3O4/g-Al2O3 NMP (10 ml) 129

�2%Co3O4/g-Al2O3 MeOH (25 ml) 215

�5%Co3O4/g-Al2O3 AcN (10 ml) 236

�5%Co3O4/g-Al2O3 DMF (10 ml) 145

�5%Co3O4/g-Al2O3 NMP (10 ml) 134

�5%Co3O4/g-Al2O3 MeOH (25 ml) 215

�8%MnO2/g-Al2O3 AcN (10 ml) 198

�8%MnO2/g-Al2O3 DMF (10 ml) 117

�8%MnO2/g-Al2O3 NMP (10 ml) 108

�8%MnO2/g-Al2O3 MeOH (25 ml) 172

a Oxidation reaction carried out at 130 8C; 25.0 ml oxidized diesel extracted

with solvent.b S content was measured by XRF and GC-AED.

3.3. Selective catalytic sulfur oxidiation followed by

solvent extraction on real diesel

Similar results were obtained with real diesel containing

approximately 450 ppm sulfur as shown by the sulfur-specific

GC-AED chromatograms in Fig. 4. The conversion of the

substituted thiophenes (Fig. 5) to corresponding sulfones was in

the range of 65–75% depending on the type of catalysts and

operating temperatures in the range of 130–200 8C. The

selectivity was about 90–100%. The total sulfur content of the

diesel before and after was same in most of the cases and when

the operating temperature increases, some of the sulfur

compounds were over oxidized and converted (see Scheme

1) into SO2 (gas). The elimination of SO2 was confirmed by

scrubbing the outlet gas with a AgNO3 solution to form AgSO3

precipitate.

Table 2 summarizes the results of extracting real diesel

before and after oxidation. Among the polar solvents tested,

NMP was found to be the most efficient in extracting sulfur

compounds from both diesel and oxidized diesel. While both

thiophenes and sulfones can be extracted from diesel, the

sulfones are significantly easier to be removed from diesel by

polar solvents due to higher polarity. The results also show that

the extraction efficiency for both thiophenes and sulfones with

the polarity of the extraction solvent.

The GC-AED carbon chromatogram shows there were no

significant changes in the product distribution before and

after oxidation of the real diesel samples. The trisubstitued

dibenzothiophenes compounds were easier to be oxidized than

the monosubstituted dibenzothiophene, such as 4-methyl-

dibenzothiophene is difficult to oxidize compared to 4,6-

diethyl-dibenzothiophene (Table 3).

Table 3

Sulfur content analysis results after solvent extraction of oxidized diesel at variou

Catalysta Reaction temperature (8C)

�2%MnO2/g-Al2O3 130

�2%MnO2/g-Al2O3 130

�2%MnO2/g-Al2O3 130

�2%MnO2/g-Al2O3 130

�2%MnO2/g-Al2O3 130

�5%MnO2/g-Al2O3 130

�5%MnO2/g-Al2O3 130

�5%MnO2//g-Al2O3 130

�5%MnO2/g-Al2O3 130

�5%MnO2/g-Al2O3 130

�8%MnO2/g-Al2O3 150

�8%MnO2/g-Al2O3 150

�8%MnO2/g-Al2O3 150

�8%MnO2/g-Al2O3 150

�11%MnO2/g-Al2O3 150

�11%MnO2/g-Al2O3 180

�13%MnO2/g-Al2O3 150

�13%MnO2/g-Al2O3 180

�5%MnO2/3%Co3O4//g-Al2O3 180

�3%MnO2/5%Co3O4//g-Al2O3 180

Oxidation reaction temperature in 8C.a 30.0 ml oxidized diesel extracted with different amount of different solvent (sib S content was measured by XRF and GC-AED.

3.4. Properties of oxydesulfurized real diesel

The treated diesel (oxidized followed with solvent extrac-

tion) was analyzed for diesel specification parameters, such as

density, cetane index, pour point, kinematic viscosity, etc. and

the results are given in Table 4. The studies show that the olefin

content of the diesel was increased and aromatic content of

the diesel was reduced substantially. Cetane index increased

s temperatures and various catalysts system

Extraction solvent (vol) S (ppm)b

NMP (10 ml � 3) 64

NMP (10 ml) 188

NMP (20 ml) 126

NMP (30 ml) 96

DMF (10 ml) 193

NMP (10 ml � 3) 51

NMP (10 ml) 168

DMF (10 ml) 187

DMF (20 ml) 145

DMF (30 ml) 115

NMP (10 ml) 143

NMP (20 ml) 119

NMP (30 ml) 103

NMP (40 ml) 61

NMP (10 ml) 104

NMP (30 ml) 66

NMP (10 ml) 84

NMP (30 ml) 44

NMP (10 ml) 109

NMP (10 ml) 123

ngle or multiple extraction).

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–93 91

Table 4

Some diesel specification analysis for the untreated and treated diesel samples

Testa Method Real diesel Treated diesel

S content (ppmw) (wt%) ASTM D3120-96 0.043 0.001

Kinematic viscosity @ 40 8C (cSt) ASTM D445-01 4.376 4.982

Density @ 15 8C (kg l�1) ASTM D4052-96 0.8541 0.8286

Olefins (vol%) ASTM D1319-99 2.4 3.6

Aromatics (vol%) ASTM D1319-99 46.4 12.5

Water content (ppm) Karl Fischer 120 139

Pour point (8C) ASTM D97-96a +6 +12

Lubricity (mm) CEC F06-A-96 175 474

Cetane index ASTM D976-91(00) 53 62.8

a Analysis carried out by Intertek Testing Services (S) Pvt. Ltd., Singapore Technical Centre.

Fig. 6. Conversion % thiophenic compounds to corresponding sulfone in real

diesel (catalyst: (A) 8%MnO2 loaded in g-Al2O3, temperature 150 8C; (B)

5%MnO2/3%Co3O4 loaded in g-Al2O3, temperature 180 8C; (C) 3%MnO2/

5%Co3O4 loaded in g-Al2O3, temperature 180 8C).

approximately by 20%. Density and other parameters were

within the required limits. Lubricity of the treated diesel was

increased by substantial amount.

3.5. Effect of the catalyst composition

Metal oxide catalysts derived from Mn, Co, W, Ni, V, Fe and

Cu, respectively, were tested for their ability to oxidize sulfur

impurities in real diesel and a model diesel mixture composed

of n-tetradecane and various substituted dibenzothiophene

compounds in air at a temperature of 110–180 8C. Among these

catalysts, only manganese and cobalt oxides were found to be

effective in catalyzing the oxidation of the thiophenes to

sulfones. The unmodified support, g-Al2O3, was also ineffec-

tive.

The effects of metal loading and reaction temperature were

investigated. Below 110 8C, the oxidation reaction was not

observed. There was no significant difference in conversions for

the oxidation of model diesel catalyzed by either 2 or 5%

Co3O4/g-Al2O3 at 130 or 150 8C. However, in the case of

MnO2/g-Al2O3 catalysts, higher metal loading led to higher

conversion at all test temperatures: 130, 150, 180 and 200 8C.

The best results were obtained with catalysts containing the

highest MnO2 loadings (11 and 13%) at 180 8C. Similar effect

was observed for the oxidation of real diesel (Table 2).

For binary mixed metal (Mn and Co) oxide catalysts, higher

activity was observed with higher Mn/Co ratio. Thus,

5%MnO2/3%Co3O4//g-Al2O3 showed better oxidation activity

than 5%Co3O4/3%MnO2//g-Al2O3 (Table 3 and Fig. 6).

3.6. Simplified process diagram for the treatment of

production S-free diesel

Fig. 7 shows a possible process diagram for the integration

of oxidative desulfurization process into an existing refinery

process unit for desulfurization of diesel. The ODS reactor unit

may be installed as downstream of a conventional hydro-

desulfurization reactor unit. Oxidative desulfurization reaction

can be carried out as a secondary desulfurization process

for diesel that have been treated by conventional HDS process.

The treated/oxidized diesel is channeled to a stirred/mixing

tank containing polar solvent for removing the oxidized sulfur

compounds. The diesel/solvent mixture is then channeled to a

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–9392

Fig. 7. Simplified process diagram for the treatment and production of S-free diesel.

settler where the treated diesel is separated from the solvent.

The solvent can be recycled by distillation and can be reused.

The treated diesel is further passed through a basic adsorbent-

bed unit for further removal of remaining sulfur-containing

compounds in the diesel. The remaining sulfones in the treated

diesel could be easily removed by adsorption compare to

thiophinic compounds.

4. Conclusion

It has been demonstrated that Mn- and Co-containing oxide

catalysts are highly effective for selective oxidation of the

refractory sulfur compounds in diesel fuel using molecular

oxygen in air at atmospheric pressure and the sulfur content can

be easily reduced to 40–60 ppm after coupled with extraction

by polar solvent. During our research studies, Murata et al. [14]

have reported recently oxidation of sulfur compounds using

molecular oxygen in the presence of cobalt catalysts and

aldehydes in monophasic system. In our system, the catalyst

(heterogeneous) can be easily reactivated and reused. The polar

solvent used for extraction can be recycled by vacuum

distillation. The low-sulfur (�10–15 ppm sulfur) diesel was

obtained by simply pass through treated diesel (oxidized and

solvent extracted diesel) into the activated basic g-Al2O3

adsorbent-bed at room temperature.

This oxidative desulfurization process has several advan-

tages over other oxidative desulfurization processes which were

reported. One advantage of this process that the reaction can be

carried out using inexpensive oxygen found air compare to

costly oxidants, such as H2O2 or ozone, which were reported in

the literature for the oxidative desuflurisation processes. In

addition, the use of air as oxidant also eliminates the need to

carry out any oxidant recovery process that is usually required if

liquid oxidants (tert-butylperoxide or H2O2) are used. Another

advantage of this process is the mild operating conditions

compared to hydrodesulfurization process which more severe

conditions are needed. Yet another advantage of this process is

the ease of integration into any existing refinery for the

production of diesel, as afforded by the mild process conditions

of liquid phase contacting and the use of air. Furthermore, the

use of a selective oxidation catalyst also permits the tuning of

experimental parameters, such as temperature and contacting

time, to achieve optimal conversion and selectivity. The

simplified process flow sheet of the oxidative desulfurization

process which can be adopted for the refineries without major

changes in the infrastructure is shown in Fig. 7.

Acknowledgements

This work was financially supported by the Agency for

Science, Technology and Research (Project No. ICES/04-

112001). J.T. Sampanthar wish to thank Prof. Hua Chun Zeng

(National University of Singapore, Singapore) and Mr. Sam

Mylvaganam (ICES) for their valuable comments and useful

discussion.

References

[1] US EPA, Regulatory Announcement: Heavy-Duty Engine and Vehicle

Standards and Highway Fuel Sulfur Control Requirements, December,

2000.

[2] C. Song, Catal. Today 86 (2003) 211;

R. Shafi, G.J. Hutchings, Catal. Today 59 (2000) 423;

K.G. Knudsen, B.H. Cooper, H. Topsoe 189 (1999) 205.

R.G. Tailleur, J. Ravigli, S. Quenza, N. Valencia, Appl. Catal. A Gen. 282

(2005) 227;

T. Viveros-Garcia, J.A. Ochoa-Tapia, R. Lobo-Oehmichen, J.A. Reyes-

Heredia, E.S. Perez-Cisneros, Chem. Eng. J. 106 (2005) 119.

[3] F. Li, P. Xu, C. Ma, Y. Zheng, Y. Qu, J. Chem. Eng. Jpn. 36 (2003) 1174.

[4] R.T. Yang, A.J. Hernandez_Maldonado, F.H. Yang, Science 301 (2003)

73;

Y. Sano, K.H. Choi, Y. Korai, I. Mochida, Appl. Catal. B Environ. 49

(2004) 219;

X.L. Ma, L. Sun, C.S. Song, Catal. Today 77 (2002) 107;

J.T. Sampanthar et al. / Applied Catalysis B: Environmental 63 (2006) 85–93 93

A.J. Hernandez_Maldonado, R.T. Yang, J. Am. Chem. Soc. 126 (2004)

992;

A.J. Hernandez-Maldonado, F. Yang, G. Qi, R.T. Yang, Appl. Catal. B

Environ. 56 (2005) 111;

P. Jeevanandam, K.J. Klabundle, S.H. Tetzler, Microporous Mesoporous

Mater. 79 (2005) 101.

[5] S. Zhang, Q. Zhang, Z.C. Zhang, Ind. Eng. Chem. Res. 43 (2004) 614;

S. Zhang, Z.C. Zhang, Green Chem. 4 (2002) 376;

A. Bosmann, L. Datsevich, A. Jess, A. Lauter, C. Schmitz, P. Wassersc-

heid, Chem. Commun. (2001) 2494.

[6] C. Li, Z. Jiang, J. Gao, Y. Yang, S. Wang, F. Tian, F. Sun, P. Ying, C. Han,

Chem. Eur. J. 10 (2004) 2277;

S. Murata, K. Murata, K. Kidena, M. Nomura, Energy Fuels 18 (2004) 116;

J. Palomeque, J.M. Clacens, F. Frgueras, J. Catal. 211 (2002) 103;

F. Zannikos, E. Lois, S. Stournas, Fuel Process. Technol. 42 (1995) 35.

[7] M. Te, C. Fairbridge, Z. Ring, Appl. Catal. A 219 (2001) 267.

[8] P.S. Tam, J.R. Kittrell, J.W. Eldridge, Ind. Eng. Chem. Res. 29 (1990) 321.

[9] F.M. Collins, A.R. Lucy, C. Sharp, J. Mol. Catal. A 117 (1997) 397;

K. Yazu, Y. Yamamoto, T. Furuya, K. Miki, K. Ukegawa, Energy Fuels 15

(2001) 1535.

[10] L.F. Ramirez-Verduzco, E. Torres-Garcia, R. Gomez-Quintana, V. Gon-

zlez_Pena, F. Murrieta-Guevara, Catal. Today 98 (2004) 289.

[11] US Patent 4,493,765.

[12] O.S. Nonaka, T. Takashima, W.H. Qian, S. Gakkaisi, J. Jpn. Pet. Inst. 42

(1999) 315.

[13] E.L. Sughrue, et al., Fish & Dollar, US Patent Application 20,040,007,501.

[14] S. Murata, K. Murata, K. Kidena, M. Nomura, Energy Fuels 18 (2004)

116.

[15] A.S. Rappas, Unipure Corporation, US Patents US 6,402,940; US 6,

406,616, Wachs WO 03/051798.

[16] J.T. Sampanthar, X. Rong, F.M. Dautzenberg, PCT Patent Application

PCT/SG2004/000160.