A vapor phase adsorptive desulfurization process for producing ultra low sulphur diesel using NiY zeolite as a regenerable adsorbent

Abstract

A NiY zeolite based vapor phase adsorptive desulfurization process has been described which can bring down sulphur concentration of a commercial BS IV grade (Euro IV equivalent) diesel from 50 ppm to a <5 ppm level. Compared to literature reports on fixed bed adsorptive desulfurization of diesel using zeolite adsorbents, the present process has the advantage of easy regenerability of the adsorbent with minimum temperature swing between adsorption and regeneration steps. Multi cycle stability of the desulfurization performance was also demonstrated.

---

1. Introduction

Diesel is the fuel of choice for heavy duty vehicles because of its better fuel economy, durability, reliability and low speed power requirement of the diesel engine. However a major concern with diesel fuel is higher emission of environmentally harmful NOx, HC, CO and particulate matter. Near zero sulphur diesel (<10 ppm sulphur) is required for effective use of advance control technologies in diesel vehicles to minimize these emissions.¹

Currently European Union has mandated 10 ppm sulphur in road transport diesel according to its Euro-V norms. In USA the Environmental Protection Agency (USEPA) has mandated 15 ppm sulphur for highway diesel. Similar sulphur specifications are being followed by other developed countries such as Canada, Japan and Korea.² India has followed a two tier system of auto fuel specification called Bharat Stage (BS). At present twenty three major cities of India are following the BS IV norm for transportation fuel which is equivalent to Euro IV and calls for 50 ppm sulphur in diesel and gasoline. Rest of the country is being supplied with BS-III fuel which contains 350 ppm sulphur in diesel and 150 ppm sulphur in gasoline. Recently Indian government has also laid a roadmap for switchover of existing 50 ppm sulphur transportation fuel to 10 ppm sulphur fuel by April 2021.³

The current practice of producing ultra low sulphur diesel is based on catalytic hydro-desulfurization (HDS) in which hydrogen gas gets consumed. The existing processes are very energy intensive when ultra low sulphur level is targeted. This is because high severity operations (high H₂ pressure, high temperature) are required to break the C–S bonds of the last trace of increasingly refractory organo-sulphur compounds present in the diesel fraction.⁴ Increased hydrogen consumption also gives a large CO₂ footprint of the process. This could be understood from the fact that one ton of hydrogen production through reforming operation in a petroleum refinery typically leads to 4–8 ton of CO₂ emission depending upon the type of hydrocarbon feed used.⁵

There is thus an impetus for developing a low cost energy efficient desulfurization process with lower process severity and hydrogen consumption. Such process can also be used as a polishing step to bring down sulphur level of commercial diesel to ultra low level. In this context adsorption based approaches appear particularly attractive due to less severity of the process scheme.^6,7

There are several reports on fixed bed adsorptive desulfurization process for ultra low sulphur diesel. Different class of adsorbents such as metal exchanged Y type zeolites,^8–14 pi acceptor functionalized resin,¹⁵ activated carbon,¹⁶ metal incorporated mesoporous silica have been tested.^17–20 Application of hierarchical beta and Y zeolite and their cerium ion exchanged form for thiophenic sulphur removal has also been reported under batch adsorption mode.^21,22 A detailed literature update on adsorptive desulfurization of transportation fuels such as gasoline, jet fuel and diesel with different class of adsorbents can also be found in a recent book chapter by Song et al.⁷ Most of these studies are focused mainly on screening of different adsorbent materials for sulphur uptake measurements from model as well as actual feed. Regeneration aspect of spent adsorbents is often a very much neglected part of these studies though it could be easily understood that adsorbent regenerability will be a crucial aspect if industrial application of an adsorptive desulfurization process has to be considered. Not much information is also available in the open literature on cyclic stability of desulfurization performance with the reported adsorbents.

In most of the studies with zeolite ^8–14 and mesoporous material ²⁰ based sorbents the adsorption was done at room temperature while regeneration was done by air oxidation at 350–450 °C. Such large temperature swing between adsorption and regeneration will be considered a major disadvantage in the context of industrial applicability of these processes. In other cases where solvent regeneration of the spent adsorbent was recommended a large solvent inventory along with suitable solvent recovery and recycle system has to be considered which will add to the complexity of the overall process. A combined solvent regeneration–thermal oxidation approach for regeneration of spent adsorbent was also suggested but this will bring additional complexity in the regeneration scheme.²³

In continuation of our earlier work ^24–26 on adsorptive desulphurization process development, in the present work we have studied feasibility of ultra deep desulfurization of a low sulphur (50 ppm) commercial diesel using a zeolite based vapor phase adsorption process. The sulphur adsorption was carried out at low pressure with negligible consumption of hydrogen. The temperature swing between the adsorption and regeneration step is much lower than the literature reported processes with similar type of adsorbent. Cyclic stability of the desulfurization performance has also been demonstrated.

2. Experimental

2.1. Adsorbent preparation and characterization

A commercial NaY zeolite (Süd-Chemie, India; Si/Al mole ratio of 2.5) powder was ion exchanged with calculated amount of 1.72 M aqueous solution of Ni(II) ion under reflux condition. Before exchange the base zeolite was calcined at 500 °C for 6 h. The amount of Ni(II) ion in the exchange solution was kept four times the theoretical cation exchange capacity of the zeolite amount taken for ion exchange. The solid material was recovered after exchange, washed thoroughly with distilled water, dried at 50 °C for 4 h and then at 90 °C for 12 h. This exchange procedure was repeated three times. After the third exchange, the dried solid obtained was calcined at 500 °C for 6 h. ICP-AES analysis of the calcined solid showed a nickel loading of 8.8 wt% which corresponds to 97% of Ni(II) exchange in the zeolite. The solid was formulated into cylindrical extrudate by using boehmite as binder. The detailed formulation procedure is reported elsewhere.²⁴

The parent NaY zeolite, the exchanged zeolites and the final alumina extrudate form were characterized with respect to their surface area, pore volume characteristics and XRD pattern.

2.2. Feed characterization

A BS IV grade diesel was procured from a commercial petrol pump in the National Capital Region, New Delhi. The feed was characterized for sulphur content by UV fluorescence technique and ASTM D86 distillation to get the boiling range. Characteristics of the commercial diesel are given in Table 1.

Table 1 Total sulphur, density and ASTM D-86 characterization of BS IV grade commercial diesel

Total sulphur by UV fluorescence	50 ppm ‘S’
Density	0.82 g/cc
ASTM D-86 distillation characteristics
IBP	188.7 °C
30% (v/v)	269.3 °C
50% (v/v)	288.6 °C
70% (v/v)	309.9 °C
90% (v/v)	341.2 °C
FBP	368.9 °C
Distillate	98.0% (v)
Residue	1.5% (v)
Loss	0.5% (v)

---

2.3. Sulphur adsorption breakthrough studies

Sulphur breakthrough experiments were carried out in a custom designed automated adsorption unit (Fig. 1). The unit has the provision of in situ thermal regeneration/activation of spent adsorbent under oxidizing and reductive atmosphere. For adsorption breakthrough experiments, 20 g of adsorbent was packed inside the stainless steel (SS 316 grade alloy) made adsorber column of 19 mm I.D. positioned inside a three zone tubular furnace. An HPLC pump was used to pump diesel feed from a reservoir to the adsorber through a liquid pre-heater which served to vaporize the feed. The diesel vapors leaving the pre-heater were mixed with a H₂ gas and entered the adsorber from the bottom. Vapor exiting the adsorber was cooled and taken to a high pressure gas–liquid (G/L) separator. After accumulation of 75 mL of condensed liquid the samples (desulphurised product) were withdrawn at different time intervals from the separator with liquid level control by a Liquid Control Valve (Badger Instruments, USA) while uncondensed vapors were vented under pressure control through a pressure control valve (Badger Instruments, USA). A constant liquid level of 75 mL was maintained in the G/L separator throughout the experiment. The samples withdrawn were analyzed for total sulphur with the help of a UV fluorescence based analyzer. Three thermocouples axially placed inside the adsorber column were used to monitor the temperatures. Prior to the first adsorption experiment the adsorber column was heated at 350 °C in the presence of flowing dry nitrogen for 2 h in order to remove any moisture present in the adsorbent bed. For liquid phase experiments the G/L separator was bypassed and products were collected directly from the adsorber outlet at different time intervals. A simplified schematic of the unit has been reported in our earlier publications.^25,26

Fig. 1 Automated fixed bed adsorption set up used in the desulfurization study.

For adsorptive desulfurization experiments the feed used was a commercial high speed diesel having 50 ppm total sulphur. Before using as feed in adsorption studies the commercial BS IV grade diesel was pretreated through a bed of porous alumina to avoid any potential fouling of the NiY adsorbent by additives particularly the metal containing ones present in the commercial diesel.²⁷ Around two liters of diesel feed was percolated through a glass column loaded with 75 g of alumina beads. There was practically no drop in sulphur content of the feed following percolation indicating zero sulphur capacity of the alumina bed. In all the vapor phase experiments adsorption pressure was kept at 4 bar absolute. The feed flow rate was kept at either at 0.5 mL min⁻¹ or at 1 mL min⁻¹ during initial experiments while the H₂ flow rate was varied in the range of 1–2 nL min⁻¹. The liquid phase experiments were carried out in absence of any H₂ flow. Here the commercial diesel feed flow was fixed at 0.5 mL min⁻¹.

2.4. Regeneration studies

After each breakthrough experiment the sulphur loaded adsorbent was regenerated in situ by a multi step thermal oxidation method. Regeneration steps and operating conditions are given below in Table 2. The regeneration procedure and conditions were optimized in our laboratory after a large number of empirical parametric studies. Spent adsorbent was regenerated by controlled oxidation with air to convert the adsorbed sulphur compounds to sulphur oxides. During oxidation a stepwise increase in air flow is required to avoid rapid temperature shoot up which may be detrimental to the adsorbent. The minimum air flow rate used was 0.1 nL min⁻¹ and the maximum was 1.0 nL min⁻¹. The flow rate was increased at every 30 minutes at an increment of 0.1 nL min⁻¹.

Table 2 Regeneration steps and operating conditions

Regeneration steps	Operating conditions
	Temperature (°C)	Pressure (bar, absolute)	Gas flow rate (nL min^−1)	Step duration (minute)
N2 purging	450	1.0	0.5	30
Air oxidation	450	1.0	0.1–1.0	300
Air cooling	450 to 380	1.0	1.0	20
N2 cooling	380 to 350	1.0	1.0	10

---

3. Results and discussion

XRD patterns of the Y zeolite in different stages of nickel ion exchange and of the final NiY-alumina extrudate is shown in Fig. 2. Characteristic XRD fingerprint of Y zeolite is present in all the samples. However a change in the intensity ratio of the 100% peak with respect to other peaks were observed in the calcined NiY sample (after 3^rd exchange) and the final alumina extrudate compared to the parent NaY and the first and second ion exchanged samples. This suggests changes in textural characteristics of the zeolite. However a comparison of the diffraction counts of the 100% zeolite peak of the parent NaY zeolite with the final calcined NiY and its alumina extrudate form suggests that the there is no loss of crystallinity of the zeolite phase and in fact the XRD counts improved slightly in the latter forms (Table 3). Compared to the parent zeolite a 23% drop in BET surface area was observed in the solid after 3^rd nickel ion exchange followed by calcination at 500 °C. After formation of the alumina extrudate there was a further 27% drop in BET surface area. The drop in surface area following high temperature nickel ion exchange can be explained due to creation of mesoporosity as is evident from the fact that ratio of mesopore/micropore area changed from 0.07 in the parent NaY to 0.24 in the NiY after the 3^rd nickel ion exchange followed by calcination. This was also corroborated by specific micropore volume data as determined by t-plot method which showed a drop of 38% from 0.24 cm³ g⁻¹ in the base NaY powder to 0.15 cm³ g⁻¹ in the NiY sample after 3^rd exchange. In the final extrudate form the corresponding value was 0.10 cm³ g⁻¹. The surface area and porosity characterization data of the parent NaY, NiY and the final extrudate form are given in Table 3.

Fig. 2 XRD patterns of the zeolite adsorbent at different stages of preparation (a) base NaY after calcination at 500 °C; (b) after 1^st Ni(II) exchange and drying at 90 °C; (c) after 2^nd Ni(II) exchange and drying at 90 °C (d) after 3^rd Ni(II) exchange and calcination at 500 °C (e) final calcined NiY-γAlumina extrudate.

Table 3 Surface area, porosity and relative X-ray crystallinity data of the parent NaY zeolite, NiY and the final alumina extrudate form^a

Sample	BET surface area (m^2 g^−1)	Mesopore/Micropore area ratio	t-Plot micropore volume (cm^3 g^−1)	XRD counts of the 100% Peak (a.u.)
a a.u.: arbitrary unit.
NaY	503	0.07	0.24	210
Calcined NiY after 3^rd Ni(II) exchange	386	0.24	0.15	220
NiY alumina extrudate	283	0.33	0.10	230

---

Effect of adsorption temperature on sulphur removal was studied by varying adsorption temperature in the range 300–400 °C in successive experiments. The operating conditions chosen were sufficient to maintain a vaporized state of diesel–H₂ mixture into the adsorber. This was established by simulating feed conditions in Aspen Plus (version 8.4). The feed vapor fraction for the all the adsorption runs under different operating conditions of feed flow rate, H₂ flow rate and temperature was simulated. The ASTM D-86 data reported in the paper was used to characterize the diesel feed. The Peng–Robinson equation of state was used in these simulations. It was established that for all feed conditions the diesel was in completely vaporized state (Table 4).

Table 4 Simulated Vapor Fraction of Feed from Aspen Plus Simulations at 4 bar

Diesel flow (mL min^−1)	H2 flow (nL min^−1)	Adsorber temperature (°C)	Feed vapor fraction
0.5	1.5	350	1.000
0.5	1.5	375	1.000
0.5	1.5	400	1.000

---

The regeneration protocol given in Table 2 was followed after each breakthrough run. The results indicate that best desulphurization performance is obtained at 350 °C temperature (Fig. 3). About 160 mL of diesel feed can be treated at 350 °C up to an effluent sulphur level of 2 ppm. While at 375 °C the effluent sulphur level was near 4 ppm after treating same amount of feed. At 400 °C the sulphur removal capacity was found to have decreased considerably when the effluent sulphur level after 160 mL feed treatment was ∼12 ppm. The trends commensurate to a physi-sorption type process where sulphur adsorption capacity of the NiY adsorbent decreases as the adsorption temperature is increased. It is to be noted that we have not observed presence of H₂S in the adsorber effluent during our vapor phase experiments suggesting absence of reactive adsorption by organo-sulphur molecules. This indicates that H₂ is not being consumed as a reactant.

Fig. 3 Effect of adsorption temperature on vapor phase desulphurisation performance (feed flow rate: 0.5 mL min⁻¹; H₂ flow rate: 1.5 nL min⁻¹).

Interestingly however when the adsorption experiments were carried out under liquid phase condition (T: 30 °C, 200 °C; P: 1 Bar) at much lower temperatures in absence of any H₂ flow, a drastic reduction in sulphur adsorption capacity was observed (Fig. 4). The effluent sulphur concentration was 10 ppm after treating about 80 mL of diesel feed when the adsorption run was carried out at 30 °C. The desulphurization performance further deteriorated when the adsorption experiment was carried out at 200 °C. The effluent sulphur level jumped to 40 ppm after treating only about 5 mL of feed and this level was maintained up to 70 mL of feed treatment after which the experiment was stopped. A drop in sulphur removal capacity with increasing temperature was also reported with NiY adsorbent under liquid phase conditions in the context of gasoline sulphur removal. Ni(II) ions in Y zeolite has been reported to participate in π-complexation type interaction with organo sulphur molecules during adsorption under liquid phase condition.²⁸ The ability of organo-sulphur molecules to form π type bond with Ni(II) ions decreases with increasing temperature.

Fig. 4 Comparison of desulphurisation performance under liquid and vapor phase adsorption conditions (feed flow rate: 0.5 mL min⁻¹; H₂ flow rate in liquid phase: 0.0 nL min⁻¹; H₂ flow rate in vapor phase: 1.5 nL min⁻¹).

In contrast to liquid phase experiments in our study, a much improved desulphurization performance was obtained under vapor phase adsorption condition at 350 °C temperature and 4 bar adsorption pressure (H₂ flow: 1.5 nL min⁻¹) where about 160 mL of diesel feed could be treated up to an effluent sulphur concentration was 2 ppm. This improved desulfurization performance could be explained by improved mass transfer kinetics in the vapor phase of the sulphur molecules to the active adsorption sites in NiY zeolite. A similar trend was also observed with NaY zeolite where equilibrium sulphur capacity as well as rate of adsorption improved with increasing temperature during adsorption of dibenzo-thiophene from a solution of hexadecane.²⁹ The result was interpreted due to improvement in mass transfer kinetics because of reduced viscosity of the solution of dibenzothiophene in hexadecane at elevated temperature. An altogether different mechanistic interpretation was given by Ko et al. and Park et al. for diesel desulphurization using nickel oxide loaded mesoporous silica SBA-15 as adsorbent.^17,18 According to these reports no sulphur removal was observed from diesel at room temperature while desulphurization ability of the adsorbent was significantly improved at higher temperatures up to 200 °C though still below the boiling range of the commercial diesel tested. These results were explained based on chemisorptions of sulphur compounds on to the NiO surface at elevated temperature. Thus the mechanism of diesel sulphur adsorption with nickel based adsorbents is also dependent on the chemical form of nickel species.

Based on above results further adsorption experiments were investigated under vapor phase condition at 350 °C under H₂ flow. H₂ atmosphere is required not only to maintain vaporized state of the diesel but also to prevent potential coke formation from diesel vapor at higher temperatures which could have a deleterious effect on adsorbent capacity by poisoning of the adsorption sites. Effect of feed flow rate on desulphurisation performance was monitored at two different feed flow rate viz. 0.5 mL min⁻¹ and 1.0 mL min⁻¹. The results are shown in Fig. 5. The run with 0.5 mL min⁻¹ was monitored up to 158 mL of effluent volume and the sulphur level gradually rose from 0.35 ppm in the initial sample to 2.6 ppm in the final sample. Deterioration in desulfurization performances was observed when the feed flow rate was increased to 1 mL min⁻¹. After 150 mL feed treatment the sulphur level in the effluent was 5.5 ppm. The results indicate a longer bed residence time of the diesel-H₂ vapor mixture facilitate adsorption of organo-sulphur molecules present in diesel. Fig. 6 shows the effect of H₂ flow rate on vapour phase desulfurization performance. The results were compared at three H₂ flow rates viz. 1.0, 1.5 and 2.0 nL min⁻¹. Other adsorption parameters such as feed flow rate, temperature and pressure were fixed at 0.5 mL min⁻¹, 350 °C and 4 bar absolute respectively. Degree of desulphurization performance follows the trend 1.5 nL min⁻¹ > 1.0 nL min⁻¹ > 2.0 nL min⁻¹. The results indicate requirement of an optimal H₂ flow rate. An optimal H₂ flow is required to maintain sufficient residence time of the H₂-diesel vapor mixture in the adsorber for effective sulphur adsorption.

Fig. 5 Effect of feed flow rate on vapour phase adsorptive desulfurization.

Fig. 6 Effect of hydrogen flow rate on vapour phase adsorptive desulfurization.

Thermal oxidative regeneration of the adsorbent bed, under air flow at 450 °C, after sulphur breakthrough experiments was monitored by an on-line gas analyzer (Horiba). Regeneration results indicate that spent adsorbent can be effectively regenerated by thermal air oxidation as indicated in Fig. 7 and 8 wherein SO₂, CO₂ and CO levels drop to very low levels at the end of the regeneration cycle.

Fig. 7 SO₂ evolution curve during oxidative regeneration [solid trace: SO₂ concentration in ppm; dotted trace: air flow in nL min⁻¹].

Fig. 8 CO and CO₂ evolution curve during oxidative regeneration.

On the basis of above parameter optimization studies further cyclic stability studies were carried out under operating conditions given in Table 5.

Table 5 Optimized adsorption conditions for multi-cycle study

Adsorbent:	NiY-Alumina Extrudate
Adsorbent amount:	20 g
Effective adsorbent bed height:	12 cm
Feed:	BS IV HSD
Total ‘S’ in feed:	50 ppm
Feed flow rate:	0.5 mL min^−1
H2 flow rate:	1.5 nL min^−1
Adsorption pressure:	4 bar(a)
Adsorption temperature:	350 °C

---

3.1. Cyclic stability studies

To test the cyclic stability of adsorbent a set of five successive sulphur breakthrough and regeneration experiments (Set-1-1 to Set-1-5) were carried out. Each sulphur breakthrough experiment was followed by oxidative regeneration of spent adsorbent under identical conditions as per the regeneration protocol given in Table 3. The sulphur breakthrough results are shown in Fig. 9. The results are fairly repeatable. The total sulphur level in the adsorber effluent was <0.7 ppm after 200 minutes of product collection. When the breakthrough run was prolonged to 420 minutes (Set-1-5), the effluent sulphur level rose to 3 ppm at the end. The average total sulphur concentration of the accumulated products up to 420 minutes of product collection time was determined to be 1 ppm. The specific breakthrough sulphur capacity of the adsorbent in our process is 0.39 mg ‘S’/g. The value closely matched with the amount of sulphur released (0.36 mg ‘S’/g) during thermal regeneration in the form of SOx during the oxidative regeneration study. Breakthrough capacity in the similar range (0.46 mg ‘S’/g) was observed by Ko et al. in their adsorption study with commercial diesel containing 186 ppm sulphur diesel feed.¹³

Fig. 9 Cyclic breakthrough curves obtained with commercial BS IV diesel feed pretreated by an alumina bed.

A further set of five runs (Set-2-1 to Set-2-5) were carried out but this time with the commercial diesel feed without treating it through the alumina guard bed. The results are shown in Fig. 10. A gradual deterioration of desulfurization performance was observed over five cycle of adsorption regeneration. This is indicative of poisoning of the active adsorption sites most probably by gradual build up of additive residues present in the commercial diesel. However more studies will be required to establish this unequivocally.

Fig. 10 Successive breakthrough curves obtained with untreated commercial BS IV grade diesel feed.

4. Conclusion

In conclusion we have showed a vapor phase adsorptive desulfurization process based on a nickel exchanged zeolite Y adsorbent. Sulphur concentration of a commercial BS IV (Euro IV equivalent) diesel containing 50 ppm sulphur could be brought down to below 5 ppm level. The process can also achieve <1 ppm sulphur level in diesel suitable for fuel cell application but with a reduced product throughput. The adsorbent was found to be regenerable under an oxidative regeneration scheme and the process showed stable desulfurization performance in multi cycle adsorption–regeneration studies when the feed was pre treated with an alumina bed. Without alumina bed pre treatment however a gradual deterioration of desulfurization performance was observed probably due to additive build up from the commercial diesel.

Acknowledgements

The authors are thankful to Mr Sanjay Kumar for his help during experimental set up and sample collection.

References

1. http://www.unep.org/transport/pcfv/pdf/publowsulfurpaper.pdf (accessed October 2014).
2. United States Environment Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12–001, March 2012, Chapter 8, Section 8.5.1, http://www.epa.gov/blackcarbon/2012report/Appendix4.pdf (accessed October 2014).
3. Auto Fuel Vision and Policy 2025, Report of the Expert Committee, Govt. of India May 2014.
4. I. V. Babich and J. A. Moulijn, Fuel, 2003, 82, 607–631.
5. S. Dasgupta, S. Bhadauria, R. Shankar, S. Divekar, A. Arya, P. Gupta, R. Chauhan, M. O. Garg and A. Nanoti, 19th Refining Technology Meet, Chennai, 12-14 November, 2014.
6. B. E. Erickson, Environ. Sci. Technol., 2003, 319 A.
7. C. Song and X. Ma, in Hydrogen and Syngas Production and Purification Technologies, ed. K Liu, C. Song and V. Subramani, AIChE-John Wiley & Sons, 2010, ch. 5, Desulfurization Technologies, pp. 219–310.
8. A. J. Hernández-Maldonado, S. D. Stamatis and R. T. Yang, Ind. Eng. Chem. Res., 2004, 43, 769–776.
9. A. J. Hernández-Maldonado and R. T. Yang, Ind. Eng. Chem. Res., 2004, 43, 1081–1089.
10. A. J. Hernández-Maldonado and R. T. Yang, AIChE J., 2004, 50, 791–801.
11. A. J. Hernandez-Maldonado and R. T. Yang, J. Am. Chem. Soc., 2004, 126, 992–993.
12. S. Velu, C. Song, M. H. Engelhard and Y. H. Chin, Ind. Eng. Chem. Res., 2005, 44, 5740–5749.
13. C. H. Ko, V. M. Bhandari, J. G. Park, J. N. Kim, S. S. Han and S. H. Cho, Prepr. Pap. - Am. Chem. Soc., Div. Fuel Chem., 2005, 50, 805–806.
14. V. M. Bhandari, C. H. Ko, J. G. Park, S. S. Han, S. H. Cho and J. N. Kim, Chem. Eng. Sci., 2006, 61, 2599–2608.
15. A. Favre-Réguillon, M. Sévignon, M. Rocault, E. Schulz and M. Lemaire, Ind. Eng. Chem. Res., 2008, 47, 9617–9622.
16. H. Jeon, C. H. Ko, S. H. Kim and J. Kim, Energy Fuels, 2009, 23, 2537–2543.
17. C. H. Ko, J. G. Park, J. C. Park, H. Song, S. Han and J. Kim, Appl. Surf. Sci., 2007, 253, 5864–5867.
18. J. G. Park, C. H. Ko, K. B. Yi, J. Park, S. Han, S. Cho and J. Kim, Appl. Catal., B, 2008, 81, 244–250.
19. C. Sentorun-Shalaby, S. K. Saha, X. Ma and C. Song, Appl. Catal., B, 2011, 101, 718–726.
20. J. Zhang, G. Qiu, L. Fan, X. Meng, Q. Cai and Y. Wang, Fuel, 2015, 153, 578–584.
21. F. Tian, X. Yang, Y. Shi, C. Jia and Y. Chen, J. Nat. Gas Chem., 2012, 21, 647–652.
22. F. Tian, Q. Shen, Z. Fu, Y. Wu and C. Jia, Fuel Process. Technol., 2014, 128, 176–182.
23. H. Sun, L. Sun, F. Li and L. Zhang, Fuel Process. Technol., 2015, 134, 284–289.
24. S. Dasgupta, V. Agnihotri, P. Gupta, A. Nanoti, M. O. Garg and A. N. Goswami, Catal. Today, 2009, 141, 84–88.
25. A. Nanoti, S. Dasgupta, V. Agnihotri, P. Gupta, A. N. Goswami, M. O. Garg, E. Tangstad, M. Stöcker, A. Karlsson and Ø. B. Vistad, Microporous Mesoporous Mater., 2011, 146, 158–165.
26. S. Dasgupta, P. Gupta, A. Arya, A. Nanoti, A. N. Goswami, M. O. Garg, E. Tangstad, Ø. B. Vistad, A. Karlsson and M. Stöcker, Fuel, 2013, 108, 184–189.
27. N. M. Ribeiro, A. C. Pinto, C. M. Quintella, G. O. da Rocha, L. S. G. Teixeira, L. L. N. Guarieiro, M. C. Rangel, M. C. C. Veloso, M. J. C. Rezende, R. Serpa da Cruz, A. M. Oliveira, E. A. Torres and J. B. de Andrade, Energy Fuels, 2007, 21, 2433–2445.
28. M. Dastanian and F. Seyedeyn-Azad, Ind. Eng. Chem. Res., 2010, 49, 11254–11259.
29. J. Jiang and F. T. T. Ng, Adsorption, 2010, 16, 549–558.

---