Published on September 5, 2016
slide 1: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 156 Adsorption of Nitrogen and Sulphur Organic-Compounds on Titania Nanotubes A. Rendon-Rivera 1 M.A. Cortes-Jacome 2 E. López-Salinas 3 M.L. Mosqueira 4 J.A. Toledo-Antonio 5 Instituto Mexicano del Petróleo Eje Central Lázaro Cárdenas 152 San Bartolo Atepehuacan G. A. Madero 07730 México D.F. México Corresponding Author e-mail: jtoledoimp.mx Abstract — Anatase TiO 2 nanoparticles and TiO 2 nanotubes were used as adsorbents to determine their selective adsorption properties on liquid-phase pure S- or N-organic compounds dibenzothiophene DBT 46- dimethyl DBT 46-DMDBT pyrrole and quinoline representatives of those contained in Diesel fuels. As well model Diesel blends added with these compounds and were tested in order to emulate a real Diesel composition. Adsorption isotherms were determined at room temperature in the case of the pure compounds and were fitted to either Langmuir or Tempkin models. In all cases TiO 2 nanotubes showed a higher adsorption performance either at breakthrough or saturation capacity. For instance at breakthrough adsorption TiO 2 nanotubes adsorbed 230 and 41 times more DBT and 46-DMDBT respectively than those in TiO 2 nanoparticles. As well at breakthrough point TiO 2 nanotubes adsorbed 22 and 7.8 times more pyrrole and quinoline respectively than those in TiO 2 nanoparticles. Saturation adsorption capacity of TiO 2 nanotubes is 1.7-1.8 times higher for S-compounds and 1.4-1.9 times higher for N-compounds than that of TiO 2 nanoparticles. In a model Diesel blend selectively N-compounds were lowered considerably 50 and 81 for quinoline and pyrrole respectively while S- compounds remained almost unchanged. These results confirm that TiO 2 nanotubes have a strong preference for N- compounds when exposed to Diesel blends having competing S-compounds. Keywords — Adsorption of organo-nitrogen organo-sulphur Sulphur Titania Nanotube adsorption properties Liquid- phase. I. INTRODUCTION Air pollution has been a matter of great concern because of the serious environmental and health problems which derived from it. An important air pollution source is the vehicle emissions produced during the combustion of fuels. In order to control these emissions the governments of many countries have established strict regulations for the content of different compounds in transportation fuels. Particularly sulphur levels are controlled to prevent deactivation of catalytic converter in gasoline vehicles eventually SO x emissions contribute to acid rain. In the US the maximum allowable sulphur content in diesel is 15 ppmw parts per million by weight meanwhile in the EU is 10 ppmw 1-6. In addition interest in ultra-low sulphur fuels is motivated by the need of using new emission-control technology and fuel cells 27-10. Oil Refineries traditionally employ hydrodesulphurization process HDS to remove organosulphur compounds from feedstocks for diesel production but achieving ultra-low sulphur levels solely on this process has become a difficult task because of the alkylsubstitued dibenzothiophenes i.e. refractory sulphur which are not easily eliminated 1112. Therefore several alternatives have been applied to improve the efficiency of HDS: catalysts activity enhancement increase of HDS process severity use of complementary non-catalytic process among others 913-16. An additional challenge that must be solved is the removal of HDS catalysts inhibitors specially nitrogen compounds which have comparable or lower reactivity than refractory sulphur species 917-20. Basic and non-basic nitrogen molecules naturally occurring in oil feedstocks for fuel production are eliminated by hidrodenitrogenation HDN however this process is significantly more difficult to carry out than HDS 1121. Both reactions occur simultaneously. so any reduction in nitrogen prior to the desulphurization will enhance efficiency of the later 2223. Denitrogenation process is also important to prevent NO x emissions upon fuel combustion 2124. The removal of organosulphur and organonitrogen molecules is a key factor to produce clean fuels that current legislation demands but new or improved technologies are needed to reach this goal. Among these technologies are oxidation 25-30 extraction 31-35 alkylation 36-38 precipitation 39-43 biodesulphurization 44 biodenitrogenation 4546 and adsorption 947. slide 2: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 157 Particularly adsorption is an attractive approach since it does not require high temperature or high pressure and little or no hydrogen is needed. 948. Several efforts have been made to understand different adsorption aspects like selectivity adsorption capacity regenerability and adsorption mechanism. All these issues depend on the adsorbent material and the characteristics of the target feedstock. Therefore diverse materials have been examined for this end in the adsorption of sulphur and nitrogen compounds contained in oil fractions. Hernandez-Maldonado et al. reported studies using zeolites with different metal cations including Ag + Cu + Ni 2+ and Zn 2+ 49-53. These ion-exchanged materials rely on a mechanism called π-complexation to selectively remove organosulphur or organonitrogen molecules from commercial fuels. Also Song and co-workers developed zeolite adsorbents for desulphurization but they suggested a mechanism based on a direct interaction between sulphur atom and the adsorption sites 954. Activated carbon is another interesting material that has been tested in adsorption due to its high specific surface area pore size distribution and surface properties which can be modified adding some functional groups 234855-57. Furthermore other researchers proposed the use of single or combined metal oxides as an option for the production of sulphur and nitrogen adsorbents. Zinc oxide was reported as the main component of the adsorbent designed for S-Zorb process although alumina silica and nickel oxide are also part of it 59. Several studies reported the use of alumina single or mixed in desulphurization and/or denitrogenation of gasoline and diesel 4485960. Adsorption of organosulphur or organonitrogen compounds on anatase TiO 2 was neglected for many years probably because the low surface area values typical of conventional anatase 100 m 2 /g these numbers being even much lower when dealing with rutile TiO 2 30 m 2 /g. Nonetheless the development of TiO 2 nanotubes resulted in materials having specific surface area of between 300-500 m 2 /g turning these materials more attractive from the adsorption and/or catalysis application viewpoint 61 62 63. The greater the surface area exposed on a given adsorbent the lower the mass or volume necessary to fill an eventual separation reservoir once in a final application. The adsorption of gaseous nitrogen and sulphur containing molecules on anatase and rutile TiO 2 surfaces were largely investigated to characterize the TiO 2 surface reactivity 64 and references therein. In general it was established that these molecules can be bonded to coordinatively unsaturated Ti sites or to fill the O-vacancies on oxygen deficient surfaces however the total amount of adsorbed sulphur was affected by surface hydroxyl group coverage and molecularly adsorbed water layer 65. Conversely the adsorption of sulphur molecules bonded to complex organic conjugated rings such as thiophene and its derivatives on TiO 2 has been scarcely explored experimentally 66. A Ti x Ce 1−x O 2 material was proposed as a promising adsorbent of sulphur molecules contained on liquid hydrocarbon fuels. Titania-based materials are encouraging as adsorbents for adsorptive desulphurization processes due to selective S−Ti binding and regenerability in oxygen/air flows. To understand the adsorption mechanism of thiophenic compounds on TiO 2 -based adsorbents for ultra-deep desulphurization of liquid hydrocarbon fuels theoretical studies of density functional theory DFT were conducted on the adsorption of thiophene on modeled anatase TiO 2 0 0 1 surface 676869. Herein we report for the first time the adsorption of sulphur and nitrogen compounds contained in light or intermediate petroleum fractions over nanostructured nanotubes and nanofibers TiO 2 . In the present work with a view to examine the influence of TiO 2 morphology adsorptive desulphurization and denitrogenation of a model diesel fuel was carried out on TiO 2 nanotubes titanate nanotubes and compared with that of a commercial anatase. The equilibrium adsorption behavior was examined using the adsorption isotherm technique on liquid phase batch adsorption experiments. As well dynamic flow adsorption in a fixed-bed system at ambient temperature and atmospheric pressure using titanate nanotubes and a commercial anatase was carried out. The adsorptive capacity on different organosulphur and organonitrogen compounds at breakthrough and saturation point were evaluated. II. EXPERIMENTAL 2.1 Adsorbents and model diesel fuels Nanotubular titanate used as adsorbent was obtained by a hydrothermal method described in 61 starting from a commercial anatase Hombikat K03 purchased from Sachtleben Chemie GmbH. Characterization of nanotubular titania were described in a previous work 61. In order to establish the influence of TiO 2 nanoparticles transformation of into nanotubes the commercial anatase was used as a reference adsorbent. The material had a crystallite size of 20 nm and specific surface area of 101 m 2 /g. Four different model diesel fuels MDF were prepared for batch adsorption experiments using sulphur or nitrogen compounds. The MD1 and MD2 model diesel fuels were prepared with dibenzothiophene DBT ≥99 or 46- dibenzothiophene 46-DMDBT 97 dissolved in hexadecane ≥99 containing 628 and 111 ppmw S respectively. The MD3 and MD4 model fuels were prepared using pyrrole 98 or quinoline 98 dissolved in a toluene 99.9– hexadecane mixture 20 and 80 vol with a 447 and 117 ppmw N respectively. All chemicals were from Sigma-Aldrich. slide 3: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 158 For fixed-bed adsorption experiments four new model diesel fuels were prepared following the above procedure. The MD5 model diesel contained DBT with a final 730 ppmw S concentration whereas MD6 fuel contained 46-DMDBT with 143 ppmw S. Model fuels containing nitrogen were prepared with pyrrole MD7 and quinoline MD8 whose final concentrations were 380 and 100 ppmw N respectively. In order to determine the adsorption selectivity a model diesel MD9 containing all the above N and S components were prepared in a 20 toluene-80 hexadecane solvent with DBT to a final 531 ppmw S 46-DMDBT 220 ppmw S pyrrole with 294 ppmw N and quinoline 131 ppmw N. 2.2 Batch adsorption experiments Liquid phase adsorptive desulphurization and denitrogenation in a batch system were measured at room temperature 20ºC and atmospheric pressure. Different dosages of the TiO 2 adsorbent were added using constant volumes of the above described model fuels then the mixtures were shaken with magnetic stirrers in order to reach equilibrium typically 0.5 h. Subsequently the supernatant liquid was extracted passed through PTFE syringe filters and stored in vials. Sulphur and/or nitrogen concentration was analyzed to measure the adsorption capacity of the titanate nanotubes. The amount of S or N adsorbed at equilibrium in milligrams per gram of solid q e was calculated according to the following equation: m C C L q e e 0 where L is the liquid fuel weight kg C o and C e are the initial and equilibrium concentrations of the solute in the liquid fuel mg/kg respectively and m is the amount of adsorbent used g. 2.3 Fixed bed adsorption experiments Dynamic adsorption experiments were carried out using the system represented in Fig. 1. The adsorption processes were performed at room temperature and atmospheric pressure using titanate nanotubes or anatase nanoparticles as adsorbents and MD5-MD8 fuels. A glass column of 3 cm i.d. and 33 cm length was packed with 2 g of adsorbent 80-100 mesh and placed in a furnace designed for this purpose. Prior to adsorption materials were heated at 300ºC for 3 h in nitrogen flow rate: 200 cm 3 /min. After this pretreatment the adsorbent was allowed to cool to room temperature and then hexadecane was introduced into the inlet of the column using a peristaltic pump at 1 cm 3 /min in order to remove any entrapped gas. Then the hexadecane head was allowed to dissipate and the feed was exposed to the model diesel with S or N compounds using the same flow rate as that of the hydrocarbon. Effluent samples were collected periodically until saturation of the adsorbent. FIG. 1. EXPERIMENTAL SYSTEM USED FOR DYNAMIC ADSORPTION PROCESSES slide 4: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 159 Breakthrough and saturation adsorption capacities were calculated using the following equation: where q is the total adsorbed S or N mg S or N/g adsorbent is the flow rate of the feed mL/min m adsorbent is the weight of the adsorbent bed g ρ fuel is the fuel density g/mL at room temperature X i is the total S or N fraction by weight in the feed C 0 is the initial S or N concentration in the feed ppmw C is the S or N effluent concentration ppmw at time t min. 2.4 Sulphur and nitrogen analysis S or N concentrations in all collected samples were measured using an Antek 9000 total S or N analyzer except for S concentration in effluent samples obtained in dynamic adsorption for which an energy dispersive X-ray fluorescence EDXRF spectrometer model S2 Ranger supplied by Bruker was employed. III. RESULTS AND DISCUSSION 3.1 Adsorption isotherms Batch mode experiments were conducted in order to explore adsorption performance of titanate nanotubes in adsorptive desulphurization and denitrogenation of model diesel fuels MD1-MD4. Experimental data were used to construct the equilibrium isotherms for each organosulphur and organonitrogen compound. The sulphur adsorption capacities q e as a function of the equilibrium concentrations C e in the liquid phase appear in Figure 2. A B FIG. 2. EQUILIBRIUM ADSORPTION ISOTHERMS OF A DBT AND B 46-DMDBT OVER TITANATE NANOTUBES. The isotherm shape indicates a favorable adsorption of the corresponding adsorbate DBT or 46-DMDBT 70. It can be observed in the plots Fig. 2 A and B that the adsorption capacity raises gradually as the equilibrium concentration of the adsorbate in the liquid phase increased. In both cases the highest added adsorbent amount belongs to the first point of the isotherm meanwhile the last point represents the minor amount of it. In these experiments the fuel volume remained constant while the varying parameter was the adsorbent dosage therefore the observed q e increase with the decrease of the adsorbent amount can be explained on the basis of adsorption sites density. As reported in an earlier work the main adsorption sites on titania nanotubes surface were OH groups 6171 then when the number of these sites decrease while keeping a constant adsorbate concentration there will be more adsorbed molecules per adsorption site resulting in higher q e values. The same behavior was observed for adsorption isotherms of nitrogen compounds whose shape also indicated a favorable adsorption. In Fig. 3 A the nitrogen adsorption capacity q e increases gradually as the equilibrium concentration of pyrrole in treated model fuel C e rose. On the other hand the first part of the quinoline adsorption isotherm Fig. 3 B shows an increase in adsorption capacity meanwhile the equilibrium concentration remained near zero indicating almost complete adsorption of this nitrogen molecule. In addition the highest and lowest dosages of titania nanotubes belong to the first and slide 5: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 160 last point of the isotherm respectively. Then the q e increase with the decrease of adsorbent dosage could also be related to adsorption sites density. It is worth mentioning that the initial toluene percentage which is an aromatic component in model diesel remained almost at the same value after adsorption suggesting a negligible interaction with adsorption sites on titanate nanotubes. This fact also indicates a non-competitive adsorption between aromatic compounds and nitrogen molecules. A B FIG. 3. EQUILIBRIUM ADSORPTION ISOTHERMS OF A PYRROLE AND B QUINOLINE OVER TITANATE NANOTUBES. To approximate the behavior of the different adsorbates on the titanate nanotubes surface three-adsorption isotherm equations were applied: Langmuir Freundlich and Temkin. The equation constants and the determination coefficients R 2 are listed in Tables 1 2 and 3 respectively. According to R 2 values Langmuir and Temkin models gave better fit to the DBT adsorption isotherm while for 46-DMDBT the best fit was only for the former model. Comparing the Langmuir isotherm parameters of the two sulphur compounds it can be noted that Q capacity of the monolayer is higher for DBT indicating a superior adsorption capacity of the titanate nanotubes with respect to this molecule 707273. However b adsorption equilibrium constant had a higher value for 46-DMDBT then a stronger interaction between this molecule and nanotubes adsorption sites OH groups occurred 707273. TABLE 1 LANGMUIR ISOTHERM PARAMETERS FOR THE REMOVAL OF SULPHUR AND NITROGEN COMPOUNDS BY TITANIA NANOTUBES. Langmuir Qmg / g b 1/ppmw R 2 DBT 7.94 0.01 0.9772 46-DMDBT 2.52 0.17 0.9746 Pyrrole 7.28 0.05 0.9029 Quinoline 5.45 1.08 0.8499 TABLE 2 FREUNDLICH ISOTHERM PARAMETERS FOR THE REMOVAL OF SULPHUR AND NITROGEN COMPOUNDS BY TITANIA NANOTUBES Freundlich K F mg 1-1/n kg 1/n / g 1/n R 2 DBT 0.748 0.382 0.8904 46-DMDBT 0.717 0.284 0.9041 Pyrrole 0.798 0.454 0.9901 Quinoline 1.979 0.299 0.9441 slide 6: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 161 For pyrrole adsorption the Freundlich model had the highest R 2 value Table 3 reflecting heterogeneous adsorption due to energetically different adsorption sites 7073. The parameter K F represents the adsorption capacity while 1/n is indicative of adsorption intensity. Values of 1/n 1 stand for favorable adsorption conditions 707273. Additionally the equilibrium adsorption data for quinoline are better described by Temkin equation in which A and B are constants. The value of B parameter denotes favorable adsorption this fact is in accordance with the isotherm shape observed on other adsorbents 73. TABLE 3 TEMKIN ISOTHERM PARAMETERS FOR THE REMOVAL OF SULPHUR AND NITROGEN COMPOUNDS BY TITANIA NANOTUBES Temkin A B kg/g R 2 DBT 1.48 0.25 0.9788 46-DMDBT 0.44 2.89 0.9548 Pyrrole 2.20 0.25 0.9532 Quinoline 0.89 15.49 0.9798 3.2 Breakthrough curves In a comparison of the dynamic adsorption process the two adsorbents titanate nanotubes and commercial anatase nanoparticles Hombikat K03 were tested in the removal of sulphur and nitrogen compounds from model diesels MD5- MD8 blends using a fixed-bed system at ambient temperature and atmospheric pressure. The adsorption capacity for the different organosulphur and organonitrogen compounds at breakthrough and saturation point were evaluated on the basis of breakthrough curves which were generated by plotting the effluent transient sulphur or nitrogen concentration C normalized by the feed C 0 versus cumulative effluent fuel volume normalized by total bed weight. TABLE 4 BREAKTHROUGH AND SATURATION ADSORPTION CAPACITIES FOR ANATASE NANOPARTICLES AND TITANATE NANOTUBES Adsorbate q breakthrough mg S or N / g adsorbent q saturation mg S or N / g adsorbent Anatase Hombikat K03 Titanate nanotubes Anatase Hombikat K03 Titanate nanotubes DBT 0.02 4.6 9.2 16.5 46-DMDBT 0.08 3.3 4.5 7.7 Pyrrole 0.10 2.2 14.4 20.8 Quinoline 0.58 4.5 5.7 10.7 MD5 was treated with anatase and titanate nanotubes respectively their breakthrough curves are shown in Fig. 4 A. It can be seen that the anatase breakthrough curve exhibited high C/C 0 values at an early stage indicating an important sulphur amount in the effluent. This performance implied a low adsorption capacity of anatase nanoparticles and also suggests a lack of strong interactions between the adsorbent and adsorbate. A different behavior was observed for titanate nanotubes breakthrough curve in which the relation C/C 0 increased gradually at higher amounts of treated fuel reflecting a larger adsorption capacity than that of anatase. In these experiments the breakthrough point was fixed at an effluent sulphur concentration of 10 ppmw and saturation obviously was reached when the C/C 0 value equals 1.0. The calculated breakthrough and saturation sulphur adsorption capacities are listed in Table 4. The obtained results clearly show that titanate nanotubes had a much better performance than anatase nanoparticles in the adsorption of DBT sulphur at both breakthrough and saturation points. The difference observed in adsorption capacity may be related to specific surface area morphology and surface chemical properties. First titanate nanotubes had higher specific surface area 320 m 2 /g 61 than anatase nanoparticles 101 m 2 /g second the morphology greatly influence the accessibility and the surface chemistry of materials. According to previous studies surface adsorption sites on TiO 2 materials changed with morphology 71 the studies demonstrated that nanotubular titania has mainly OH groups on its surface while anatase nanoparticles is characterized by surface Ti 4+ cations 71. Also nanotubes used in this work had only OH groups at their surface as reported in a previous work 61. On the basis of these different results the higher adsorption capacity found in titanate nanotubes suggests that DBT had a stronger interaction with OH groups than with Ti 4+ . Furthermore said interaction might be via the formation of slide 7: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 162 H-bonds between the proton of OH groups and S atom in DBT molecule since non-Brönsted nature prevails in OH groups 71. The model diesel fuel prepared with 46-DMDBT MD6 was also treated with titanate nanotubes and anatase nanoparticles. The breakthrough curves for 46-DMDBT sulphur adsorption are shown in Fig. 4 B. With anatase nanoparticles breakthrough curve rapidly reached C/C 0 values over 0.08 and saturation was clear at a very early stage the breakthrough point fixed at 10 ppmw S was exceeded from the beginning. Breakthrough curve behavior indicated a low 46-DMDBT sulphur adsorption capacity and also suggests the existence of very weak interactions between adsorbent and adsorbate. On the other hand titanate nanotubes breakthrough curve showed a slow C/C 0 increase at larger volumes of treated fuel up to saturation reflecting a superior sulphur adsorption capacity. Table 4 shows the breakthrough and saturation sulphur adsorption capacities which validate a better performance of titanate nanotubes in the adsorption of 46-DMDBT. As in the case of DBT the higher adsorption capacity of nanotubes could be related to their specific surface area morphology and nature of surface sites. Aside from having different surface adsorption sites on titanate nanotubes and anatase nanoparticles the accessibility to these sites could play an important role in the adsorption of 46-DMDBT. As reported elsewhere by our group 71 nanotubular titania have more accessible adsorption sites OH groups therefore a stronger interaction can be expected between 46-DMDBT and protons exposed on the curved nanotubular surface than with adsorption sites Ti 4+ cations located at nanoparticles flat surface. The steric hindrance conferred by the two methyl groups located in 4 and 6 positions of 46-DMDBT reduces the interaction with a protruding proton on a curved nanotubular structure. A B FIG. 4. BREAKTHROUGH CURVES OF A DBT AND B 46-DMDBT OVER ■TITANATE NANOTUBES AND ●ANATASE NANOPARTICLES. In a comparison of DBT and 46-DMDBT sulphur adsorption capacity on nanotubes the lower value obtained for the latter compound suggests that steric hindrance might play an important role in adsorption process although it was reduced to certain extent due to the accessibility of OH groups located on the curved nanotubular surface. This result also suggests a direct interaction between S atom of the model organosulphur compounds and the proton of OH groups that may occur via the formation of H-bonds. The literature on adsorptive desulphurization reported the performance of various adsorbents using model and commercial diesel. For instance Song et al. desulphurized model diesel using activated carbon derived from coconut and wood obtained breakthrough and saturation adsorption capacities of approximately 6.5 and 16.7 mg S/g adsorbent respectively for the former and 4.3 and 13.1 mg S/g adsorbent for the latter 3. Also Mochida et al. removed sulphur using activated carbon from a hydrotreated gasoil for which the breakthrough adsorption capacity was reported to be approximately 10 mg S/g adsorbent 74. On the other hand also a successful group of adsorbents was developed by Hernández-Maldonado and co- workers. They synthesized ion exchanged zeolites to remove not only sulphur but nitrogen compounds from commercial diesel the highest total sulphur adsorption capacity reported was approximately 8.89 and 13.12 mg S/g adsorbent at breakthrough and saturation respectively for CuI-Y zeolite 4950. Lower total sulphur adsorption capacities were reported for other ion-exchanged zeolites: CuI-Y 5.34 and 11.96 mg S/g adsorbent NiII-X 4.57 and 8.03 mg S/g adsorbent NiII-Y 5.05 and 9.24 mg S/g adsorbent CeIV-Y 1.37 and 3.90 mg S/g adsorbent the former values belong to slide 8: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 163 the breakthrough and the latter to saturation 5075. Furthermore Hernández-Maldonado et al. increased the sulphur removal making use of guard beds of activated carbon and/ or alumina in combination with zeolite adsorbents 5176. However they observed that the amount of aromatics played an important role in diminishing the sulphur removal capacity 76. The values of sulphur adsorption capacity reported for titanate nanotubes in this study are in the range of that obtained for some ion-exchanged zeolites although the model fuel only contained DBT or 46-DMDBT as sulphur compounds. This fact indicates that titanate nanotubes may be a promising adsorbent for desulphurization of fuels. A B FIG. 5. BREAKTHROUGH CURVES OF A PYRROLE AND B QUINOLINE OVER □TITANATE NANOTUBES AND ○ANATASE NANOPARTICLES. Additionally our adsorbents were tested in denitrogenation of MD7 and MD8 which contained pyrrole or quinoline respectively. The breakthrough curves for the adsorption of pyrrole over anatase and nanotubes are shown in Fig. 5 A. It can be observed that for anatase nanoparticles the C/C 0 values increased gradually indicating a moderate adsorption of pyrrole molecules however the breakthrough point fixed in 10 ppmw N was overtaken from the begging of the process. A better adsorption performance was seen for nanotubes breakthrough curve in which C/C 0 values also increased gradually but at higher volumes of treated fuel. The calculated breakthrough and saturation adsorption capacities are listed in Table 4 in which nanotubes have larger values as expected from the behavior of breakthrough curves. It is worth mentioning that the percentage of toluene in untreated model fuel remained almost the same after adsorption process indicating a minor adsorption of this aromatic compound if any in both anatase and nanotubes. This result agrees with that obtained in batch adsorption. As mentioned above in the adsorption of organosulphur molecules a better performance of nanotubes in the adsorption of pyrrole can be attributed to physicochemical characteristics. The larger value of nanotubes specific surface area morphology the type and the accessibility of adsorption sites might contribute to enhance adsorption. The obtained results also suggest a stronger interaction between OH groups in nanotubes and pyrrole molecules than with anatase´s Ti 4+ cations. Finally the breakthrough curves in the adsorption of quinoline over anatase and titanate nanotubes are presented in Fig. 5 B in which a superior performance of nanotubes can be observed. In anatase breakthrough curve the C/C 0 values increased progressively reflecting a moderate quinoline adsorption up to saturation. The adsorption of quinoline over anatase might be due to an acid-base interaction between the lone electron pair located in nitrogen atom and Ti 4+ cations that exist at the surface of the adsorbent since in our previous work said interaction was clearly revealed on pyridine and lutidine adsorption over anatase nanoparticles 71. On the other hand it is clear that a larger capacity for nitrogen adsorption can be obtained on titanate nanotubes in view of the fact that around 65 mL of fuel with a N concentration lower than 10 ppmw were produced as showed by breakthrough curve. After breakthrough point fixed at 10 ppmw the C/C 0 value increased gradually reaching saturation at a treated fuel amount of 112 mL. The breakthrough and saturation adsorption capacities for both adsorbents are indicated in Table 4 which confirmed the superior performance of titanate nanotubes. Also in these experiments the initial toluene percentage contained in model fuel remained almost at the same value after adsorption. The better performance of nanotubes in quinoline nitrogen adsorption again can be related to their specific surface area morphology and type of adsorption sites OH groups. According to pyridine and lutidine adsorption on titania nanotubes slide 9: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 164 reported previously by our group 71 the main interaction between these adsorbates and adsorbent was via H-bonding therefore it is expected that the same interaction had occurred between the lone electron pair located in N atom of quinoline and OH groups which are titanate nanotubes adsorption sites. Comparing the pyrrole and quinoline nitrogen adsorption capacity of nanotubes a larger value was obtained at breakthrough point for the latter compound meanwhile at saturation the higher value was for the former. This behavior reflects a higher affinity for quinoline than for pyrrole. Fewer results for denitrogenation of model and commercial diesel can be found in literature than for desulphurization ones. Mochida et al. studied adsorptive performance of activated carbon fibers in denitrogenation of a gasoil and the obtained capacity was 5.5 mg N/g adsorbent 74. As well Hernández-Maldonado et al. reported the use of CuI-Y zeolite adsorbent combined with an activated carbon guard bed in denitrogenation of a commercial diesel for which an adsorption capacity at breakthrough point of approximately 3 mg N/g adsorbent was obtained 53. Titanate nanotubes tested in this study presented a similar and superior adsorption capacity for nitrogen compounds according to the values reported in Table 4 furthermore a minor adsorption of aromatic toluene was observed during denitrogenation indicating a negligible competitive effect if any for adsorption sites of titanate nanotubes between heteronitrogen molecules and aromatics. In order to establish the selectivity of adsorbent towards nitrogen or sulphur compounds a model diesel MD9 containing both nitrogen and both sulphur compounds were prepared and contacted with the titanates nanotubes at weight/volume ratio of 0.05. As can be seen in Table 5 nitrogen compounds were selectively adsorbed on the nanotubes. Around 81 and 50 of pyrrole and quinoline respectively were eliminated from MD9 model diesel whereas both sulphur compounds practically remained unchanged after adsorption process aiming that TiO 2 nanotubes absorb preferentially nitrogen compounds. TABLE 5. ADSORPTION SELECTIVITY OF TITANATE NANOTUBES ON N OR S ORGANO-COMPOUNDS CONTAINED IN A MODEL DIESEL Adsorbate MD9 ppmw MD9 after adsorption ppmw Amount of N or S adsorbed DBT 531 525 1.0 46-DMDBT 220 220 0 Pyrrole 294 56 81 Quinoline 131 65 50 IV. CONCLUSIONS Titanate nanotubes are a promising adsorbent of organosulphur and organonitrogen compounds contained in diesel fuel according to our results obtained in equilibrium and dynamic adsorption tests. Equilibrium adsorption experiments revealed favorable adsorption of sulphur DBT 46-DMDBT and nitrogen pyrrole quinoline molecules. The DBT adsorption experimental data fitted to a Langmuir and Temkin models while for 46-DMDBT the Langmuir model had the best fit. Calculated parameters of Langmuir isotherm indicated that titanate nanotubes adsorbed higher higher amounts of DBT however 46-DMDBT had a stronger interaction with them. In the case of nitrogen compounds the Freundlich model fit better the pyrrole adsorption isotherm reflecting heterogeneous adsorption due to energetically different adsorption sites meanwhile Temkin model better describe quinoline equilibrium adsorption data which takes into account some indirect interactions between adsorbate and adsorbent. In fixed-bed adsorption experiments titanate nanotubes had a much better performance than anatase nanoparticles in the adsorption of sulphur and nitrogen compounds. The difference observed in adsorption capacity may be related to specific surface area morphology and/or surface chemical properties such as surface functional group type. Also the accessibility of the adsorption sites might play an important role especially in the adsorption of molecules with steric hindrance as in 46- DMDBT adsorption capacity values for anatase nanoparticles and titanate nanotubes. Particularly the results of sulphur compounds adsorption on titanate nanotubes suggest the existence of a direct interaction between the heteroatom of the adsorbate and the surface OH groups in the adsorbent. Results of fixed-bed nitrogen adsorption pointed out nanotubes have higher affinity for quinoline than for pyrrole since at breakthrough point the adsorption capacity was higher for the former compound. According to previously reported slide 10: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 165 experiments 71 which suggest that the main interaction between quinoline and OH groups on titanate nanotubes occurred via H-bonding. In addition a negligible uptake of toluene on titanate nanotubes was observed in equilibrium and dynamic adsorption experiments. This fact represents an advantage of the nanotubes in desulphurization and denitrogenation of diesel fuels which contain high amounts of aromatic compounds. ACKNOWLEDGEMENTS Adriana Rendón Rivera is indebted to CONACYT for a graduate student scholarship. The authors acknowledge financial support from Instituto Mexicano del Petróleo Projects D.00246 and D.00483. REFERENCES 1 K.G. Knudsen B.H. Cooper H. Topsøe Appl. Catal. A: Gen. 189 1999 205-215. 2 X. Ma L. Sun C. Song Catal. Today 77 2002 107-116. 3 A. Zhou X. Ma C. Song J. Phys. Chem. B 110 2006 4699-4707. 4 J. Hyung Kim X. Ma A. Zhou C. Song Catal. Today 111 2006 74-83. 5 J.M. Kwon J.H. Moon Y.S. Bae D.G. Lee H.C. Sohn C.H. Lee ChemSusChem 1 2008 307-309. 6 M. Muzic K. Sertic-Bionda Z. Gomzi S. Podolski S. Telen Chem. Eng. Res. Des. 88 2010 487-495. 7 X. Ma M. Sprague C. Song Ind. Eng. Chem. Res. 44 2005 5768-5775. 8 Z. Cheng X. Liu J. Lu M. Luo React. Kinet. Catal. Lett. 97 2009 1-6. 9 C. Song Catal. Today 86 2003 211-263. 10 C. Shalaby X. Ma A. Zhou C. Song Energy Fuels 23 2009 2620-2627. 11 B.C. Gates H. Topsøe Polyhedron 16 1997 3213-3217. 12 H. Rang J. Kann V. Oja Oil Shale 23 2006 164-176. 13 M. Breysse G. Djega-Mariadassou S. Pessayre C. Geantet M. Vrinat G. Pérot M. Lemaire Catal. Today 84 2003 129-138. 14 M.F. Ali A. Al-Malki S. Ahmed Fuel Process. Technol. 90 2009 536-544. 15 R. Shafi G.J. Hutchings Catal. Today 59 2000 423-442. 16 F. van Looij P. van der Laan W.H.J. Stork D. J. DiCamillo J. Swain Appl. Catal. A 170 1998 1-12. 17 H. Schulz W. Böhringer P. Waller F. Ousmanov Catal. Today 49 1999 87-97. 18 G.C. Laredo S. J.A. De los Reyes H. J.L. Cano D. J. Castillo M. Appl. Catal. A 207 2001 103-112. 19 K.H. Choi Y. Korai I. Mochida J.W. Ryu W. Min Appl. Catal. B 50 2004 9-16. 20 P. Zeuthen K.G. Knudsen D.D. Whitehurst Catal. Today 65 2001 307-314. 21 P. Wiwel K. Knudsen P. Zeuthen D. Whitehurst Ind. Eng. Chem. Res. 39 2000 533-540. 22 H. Yang J. Chen Y. Briker R. Szynkarczuk Z. Ring Catal. Today 109 2005 16-23. 23 Y. Sano K.H. Choi Y. Korai I. Mochida Energy Fuels 18 2004 644-651. 24 H. Zhang G. Li Y. Jia H. Liu J. Chem. Eng. Data 55 2010 173-177. 25 J.M. Campos-Martin M.C. Capel-Sanchez J.L. G. Fierro Green Chem. 6 2004 557-562. 26 J. Zhang W. Zhu H. Li W. Jiang Y. Jiang W. Huang Y. Yan Green Chem. 11 2009 1801-1807. 27 W. Zhu H. Li X. Jiang Y. Yan J. Lu J. Xia Energy Fuels 21 2007 2514-2516. 28 G. Yu S. Lu H. Chen Z. Zhu Energy Fuels 19 2005 447-452. 29 Y. Shiraishi K. Tachibana T. Hirai I. Komasawa Ind. Eng. Chem. Res. 41 2002 4362-4375. 30 P. Q. Yuan Z. M. Cheng X.Y. Zhang W.K. Yuan Fuel 85 2006 367-373. 31 S. Kumar J. Gentry Hydrocarbon Engineering 8 2003 27-28. 32 J. Qi Y. Yan W. Fei Y. Su Y. Dai Fuel 77 1998 255-258. 33 S. Zhang Q. Zhang Z.C. Zhang Ind. Eng. Chem. Res. 43 2004 614-622. 34 H. Li W. Zhu Y. Wang J. Zhang J. Lu Y. Yan Green Chem. 11 2009 810-815. 35 E.S. Huh A. Zazybin J. Palgunadi S. Ahn J. Hong H.S. Kim M. Cheong B. S. Ahn Energy Fuels 23 2009 3032-3038. 36 B.D. Alexander G.A. Huff V. R. Pradhan W.J. Reagan R.H. Cayton US Patent 6024865 2000. 37 N.A. Collins J.C. Trewella US Patent 5599441 1997. 38 B.H. Olivier D. Uzio F. Diehl M. Lionel FR Patent 2840916 A1 2003. 39 V. Meille E. Schulz M. Vrinat M. Lemaire Chem. Commun. 1998 305-306. 40 A. Milenkovic E. Schulz V. Meille D. Loffreda M. Forissier M. Vrinat P. Sautet M. Lemaire Energy Fuels 13 1999 881-887. 41 Y. Shiraishi Y. Taki T. Hirai I. Komasawa Ind. Eng. Chem. Res. 40 2001 1213-1224. slide 11: International Journal of Engineering Research Science IJOER ISSN: 2395-6992 Vol-2 Issue-8 August- 2016 Page | 166 42 Y. Shiraishi K. Tachibana T. Hirai I. Komasawa Ind. Eng. Chem. Res. 40 2001 3390-3397. 43 Y. Shiraishi K. Tachibana T. Hirai I. Komasawa Ind. Eng. Chem. Res. 40 2001 4919-4924. 44 S.C.C. Santos D.S. Alviano C.S. Alviano M. Pádula A.C. Leitão O.B. Martins C.M. Ribeiro M.Y.M. Sassaki C.P.S. Matta J. Bevilaqua G.V. Sebastián L. Seldin Appl. Microbiol. Biotechnol. 71 2006 355-362. 45 M.J. Benedik P.R. Gibbs R.R. Riddle R.C. Willson Trends Biotechnol. 16 1998 390-395. 46 S. Le Borgne R. Quintero Fuel Process. Technol. 81 2003 155-169. 47 A.J. Hernández-Maldonado R. T. Yang Catal. Rev.: Sci. Eng. 46 2004 111-150. 48 M. Almarri X. Ma C. Song Ind. Eng. Chem. Res. 48 2009 951-960. 49 A.J. Hernández-Maldonado R. T. Yang J. Am. Chem. Soc. 126 2004 992-993. 50 A.J. Hernández-Maldonado F. H. Yang G. Qi R. T. Yang Appl. Catal. B 56 2005 111-126. 51 A.J. Hernández-Maldonado S.D. Stamatis R.T. Yang A.Z. He W. Cannella Ind. Eng. Chem. Res. 43 2004 769-776 52 R.T. Yang A. Takahashi F.H. Yang Ind. Eng. Chem. Res. 40 2001 6236-6239. 53 A.J. Hernández-Maldonado R.T. Yang Angew. Chem. Int. Ed. 43 2004 1004-1006. 54 X. Ma S. Velu J. Hyung Kim C. Song Appl. Catal. B 56 2005 137-147. 55 Y. Sano K. H. Choi Y. Korai I. Mochida Appl. Catal. B 49 2004 219-225. 56 A. Zhou X. Ma C. Song J. Phys. Chem. B 110 2006 4699-4707. 57 A. Zhou X. Ma C. Song Appl. Catal. B 87 2009 190-199. 58 I.V. Babich J.A. Moulijn Fuel 82 2003 607-631. 59 J. Wu X. Li W. Du C. Dong L. Li J. Mater. Chem. 17 2007 2233-2240. 60 A. Srivastav V.C. Srivastava J. Hazard. Mater. 170 2009 1133-1140. 61 A. Rendón-Rivera J.A. Toledo-Antonio M.A. Cortés-Jácome C. Angeles-Chávez Catal. Today 166 2011 18-24. 62 J.A. Toledo-Antonio S. Capula M.A. Cortes-Jacome C. Angeles-Chavez E. Lopez-Salinas G. Ferrat J. Navarrete J. Escobar J. Phys. Chem. C 111 2007 10799-10805. 63 T. Kasuga M. Hiramatsu A. Hoson T. Sekino K. Niihara Langmuir 14 1998 3160-3163. 64 U. Diebold Surf. Sci. Rep. 48 2003 53-229 65 J. Baltrusaitis P.M. Jayaweera V.H. Grassian J. Phys. Chem. C 115 2011 492–500 66 S. Watanabe X. Ma C. Song Prepr. Pap. –Am. Chem. Soc. Div. Fuel Chem. 49 2 2004 511-513. 67 S. Sitamraju M.J. Janik C. Song Top Catal 55 2012 229–242. 68 J. Guo M. J. Janik C. Song J. Phys. Chem. C 116 2012 3457−3466. 69 J. Guo S. Watanabe M.J. Janik X. Ma C. Song Catal. Today 149 2010 218–223. 70 F.L. Slejko Adsorption technology: a step by step approach to process evaluation and application Marcel Dekker New York 1985. 71 J.A. Toledo-Antonio M.A. Cortés-Jácome J. Navarrete C. Angeles-Chávez E. López-Salinas A. Rendón-Rivera Catal. Today 155 2010 247-254. 72 J.D. Seader E.J. Henley Separation Process Principles John Wiley Sons United States of America 1998. 73 B.H. Hameed D.K. Mahnoud A.L. Ahmad J. Hazard. Mater. 185 2008 65-72. 74 Y. Sano K. Sugahara K.H. Choi Y. Korai I. Mochida Fuel 84 2005 903-910. 75 A.J. Hernández-Maldonado R. T. Yang Ind. Eng. Chem. Res. 43 2004 1081-1089. 76 A.J. Hernández-Maldonado R. T. Yang Ind. Eng. Chem. Res. 42 2003 3103-3110.