SciELO - Scientific Electronic Library Online

 
vol.61 issue4ELECTROCHEMICAL METHOD FOR SULFITE DETERMINATION IN WINES BY ELECTROCHEMICAL RESPONSE USING A MEMBRANE ABSORBER SYSTEMIMPLEMENTATION OF AN ANALYTICAL METHOD FOR THE DETERMINATION OF INORGANIC ARSENIC SPECIES IN OCCUPATIONALLY EXPOSED HUMAN URINE SAMPLES AND ITS TOXIC EFFECTS ON EPITHELIAL CELLS OF RENAL COLLECTING TUBULE author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Journal of the Chilean Chemical Society

On-line version ISSN 0717-9707

J. Chil. Chem. Soc. vol.61 no.4 Concepción Dec. 2016

http://dx.doi.org/10.4067/S0717-97072016000400012 

 

DENITROGENATION BY ADSORPTION OF PYRIDINE ON NI/SUPPORT ADSORBENTS

 

CECILIA PERALTA1, ESTEBAN CAMÚ1, R. BASSI1, MIRZA VILLARROEL2, JUAN OJEDA3, PATRICIO BAEZA1

1 Pontificia Universidad Católica de Valparaíso, Instituto de Química, Casilla 4059, Valparaíso, Chile.
2 Universidad de Santiago de Chile, Facultad de Química y Biología, Casilla 40, Correo 33, Santiago, Chile.
3 Universidad de Valparaíso, Facultad de Farmacia, Casilla 5001, Valparaíso, Chile.
* e-mail: patricio.baeza@pucv.cl


ABSTRACT

In this work we have carried out the adsorption of pyridine using three different supports (activated carbon, SiO2 and y-Al2O3). After choosing the best support, due to its higher adsorption capacity, we have impregnated the support with nickel at three different concentrations (2, 4 and 6% w/w) by wet impregnation to study the adsorption of pyridine by π-complexation. All the samples (supports and adsorbents) were characterized by N2 adsorption-desorption by the BET method and electrophoretic migration. The experimental results, for the three different supports, show that the adsorption capacity is better for y-Al2O3 due to its higher isoelectric point. With the incorporation of nickel, no better adsorption capacities are observed, due that the nickel incorporation diminish the zero point charges of the adsorbents.

Keywords: Adsorption, denitrogenation, nickel, pyridine.


 

1. INTRODUCTION

Crude oil is the largest and most widely used source of energy in the world and projections assure that this will continue until at least 2050. Overexploitation of existing fields, has led to crude oils known as "Barrel Fund", composed of high molecular weight hydrocarbons which are of lower quality, as they are much heavier and therefore difficult to purify in the petroleum refining industry. The combustion of petroleum containing nitrogen compounds emits NOx into the atmosphere, which directly contributes to the production of acid rain and Photochemical Smog. Therefore, environmental regulations are becoming more stringent with regard to the maximum level of nitrogen molecules that petroleum may contain; this has caused refiners to seek new methods that allow them to eliminate nitrogenous molecules. These nitrogen containing molecules are more resistant to the existing process used in the refining industry to remove them. This hydrotreating process is called hydrodenitrogenation (HDN) and it operates at elevated temperatures and pressures, which are efficient in removing simple nitrogen molecules, but not for the removal of "refractory molecules". Nitrogenous compounds present in the petroleum may vary depending on the origin of the oil field. Despite this, the predominant compounds consist of substituted pyrrole and pyridine, which containing aromatic rings of 5 and 6 members, respectively1. At present, petroleum is composed of alkylated nitrogenous organic molecules, on some part of the aromatic ring, which are refractory to the HDN process2,3.

It should be noted that the catalysts used in petroleum refining for the HDN reaction are Ni-Mo/y-Al2O3 or Co-Mo/y-Al2O3 catalysts, where Co and Ni promote the activity of MoS2, but an important point is that these catalysts are also active in the hydrodesulfurization (HDS) reaction, and furthermore, both reactions are carried out in the same reactor with an unique catalyst, and in the case of nitrogen containing compounds, their removal is highly recommended to ensure successful HDS. It is these nitrogen compounds that inhibit HDS and other hydrotreating processes, because these molecules are preferentially adsorbed at the same catalytic site3,4.

The most common methods for the removal of nitrogen compounds are biological degradation5-7, photocatalysis8, ozonation9, membrane separation10, and adsorption11-14. It is interesting to find and to study new alternatives to remove or separate the nitrogenous compounds from petroleum in order to reduce the NOx generated in the combustion of petroleum and diminish the magnitude of the inhibitory effect of these molecules in other petroleum refining processes; implementing new purification processes that are more environmentally friendly.

The use of adsorption to selectively remove the nitrogen compounds by a π-complexation mechanism under ambient conditions is an excellent option, the metallic cation can form the usual σ bonds with their s-orbitals, while their d-orbitals can back-donate electron density to the antibonding π-orbitals of the nitrogen rings in organic nitrogen compounds. The metals that can form strong π-complexation bonding are those that possess empty s-orbitals and the necessary electron density available at the d-orbitals for back donation, thereby, Cu1+, Ag1+ and Ni2+ are good options; within the above mentioned cations, Ni2+ in oxidized state is simply obtained by a calcination process15-17. In previous studies where the adsorption of 4,6-DMDBT was studied on activated carbon impregnated with different metals of the first transition series18, it was found that the adsorption capacity of the materials depends on the nature of the metal which acts as active site in the n type adsorption and we observed a volcano type curve with a maximum in Ni with oxidation state +2.

It s well known, that activated carbon has excellent adsorption properties, due mainly to its high specific area19, but also, other different materials have been investigated as adsorbents, in the case of alumina, It has been used in the removal of organic compounds from aqueous solutions20 as well as in removing phosphates21 and heavy metals22,23, these good adsorption properties are due mainly to specific area of the alumina and the amount of hydroxyl groups present on its surface. Silica has been studied in the denitrogenation by adsorption of a raw diesel24, and light gas oil (LGO)25 , among other adsorption studies26,27, so, it will be interesting to study which of the three support possess a better adsorption performance, to impregnate it with nickel at 2, 4, 6% w/w and evaluate the n-complexation mechanism, using pyridine as a model molecule.

2. EXPERIMENTAL

2.1. Preparation of the adsorbent

The supports studied were: SiO2, y-Al2O3 and activated carbon (C), these commercial supports were used in a particle size between 120 and 270 mesh, this samples were dried at 100 °C for 2 hours in a muffle to remove traces of water.

The adsorbents were prepared by dry impregnation of the best support, obtained in support adsorptions studies, with an aqueous solution of Ni(NO3)2x6H2O, containing the required amount of precursor salt to render a nominal concentrations of Ni at 2, 4 or 6% w/w. After impregnation of the supports, the adsorbents were dried at 100 °C for 12 h, and then calcined at 300 °C for 3h in a muffle furnace.

2.2. Samples characterization

For specific area (BET) determination 0.2-0.4 g of each sample were taken, which were degassed at 120 °C for 18 hours, reaching a final pressure of 1x10-3 mmHg. Subsequently, the analysis was performed within a glass slide at -196 °C using a Micromeritics ASAP-2010 apparatus for volumetric nitrogen adsorption-desorption.

Zeta potential measurements by electrophoretic migration were performed on a Zeta Meter, model 3.0, provided with an automatic transfer unit sample. In each measurement is used approximately 30 mg of the sample suspended in 300 mL of KCl solution 1x10-3 M. Between each point, the pH was adjusted with 0.1 M KOH or HCl as required.

2.3. Adsorption experiments

Adsorption experiments were performed in a vertical Pyrex reactor equipped with a supporting glass porous disk. A bed of ~500 milligrams of different adsorbents was placed in the reactor. Prior to the determination of the adsorption capacity (Ac), a process was performed for each support and adsorbent, with a 20 mL min-1 flow of pure Ar at 200°C for 30 min and subsequent cooling. Finally, the temperature was lowered again to ambient temperature with a 20 mL min-1 flow of pure Ar.

For the adsorption experiments, the liquid flow was driven into the reactor by means of a variable flow peristaltic mini-pump. The experiment was performed using an artificial mixture of 500 ppmv of pyridine in isooctane at a feed rate of ~0.5 mL min-1. Samples were collected every 5 min, until saturation of the adsorbent was achieved; the total time and number of samples depended on the amount of adsorbent and its adsorption capacity.

The samples were analyzed by gas chromatography with a flame ionization detector (GC-FID), using a Shimadzu GC-2010 equipped with an SPB-5 capillary column (L 30m, I.D. 0.25 mm, Film 0.25 mm), and the following conditions: detector at 250 °C, injector at 150 °C, and carrier flow, 15 mL min-1 (N2). The column temperature was set to increase from 160 °C to 200 °C at a rate of 40 °C min-1, and then at 220 °C at 5 °C min-1; 0,5 pL of the sample volume was injected, using an autosampler, for each GC-FPD run. The limit detection was 2 ppm.

3. RESULTS AND DISCUSSION

3.1. Effect of the nature of support

Characterization results by the BET method and isoelectric points by Electrophoretic Migration measurements of the different supports studied are shown in Table 1.

 

Table 1: Specific BET area, IEP and adsorption capacities
of Pyridine per grams of different supports studied.

 

It is clear that the Cact have a larger specific area, it is also noted that y-Al2O3 is the support that has the smallest pore volume, also y-Al2O3 has the highest isoelectric point of the supports studied (8.7), SiO2 and C have very similar isoelectric points, 3.2 and 4.0 respectively; It should be noted that this causes that the supports studied having negative charge at pH > IEP. Note that for the case of alumina to pH < 8.7, will have a net positive charge; C and SiO2 may have net positive charge only at very acidic pH.

By integrating the area under the adsorption curves, the saturation adsorption capacities can be estimated. The Adsorption capacities of pyridine for supports studied are shown also in Table 1. Firstly, it is clear that there is no relationship between the specific areas of the supports studied and its adsorption capacity.

The results shown that within the supports studied, y-Al2O3 is the only support having a substantial capacity to adsorb pyridine between the studied support (22.1 mg (g of support)-1); this is mainly due to that the feed used has a pH near 7.0, less than the IEP of alumina (8.7), at pH 7.0 the alumina possesses a net positive charge or electron deficient so that the pyridine adsorption is favored, because pyridine is a Lewis base; on the other hand, silica and Cact have a minor IEP (3.2 and 4.0, respectively) and at pH 7.0 both supports have negative net charge or excessive electrons pairs, preventing the adsorption of the pyridine as Lewis base. This behavior can be seen more clearly in Fig.1, in which the dependence of the adsorption capacity of the materials studied with its IEP is observed, clearly shows a decrease in the adsorption capacity as the IEP of the support is less than 7.0; in the case of C this has a slight adsorption capacity which can be explained by the presence of a surface group that may have a pH 7.0 acidic property and can be protonated at the pH; or the large specific surface area of this material.

 

Fig. 1 Adsorption capacities of Pyridine (bars) and IEP
(dots) for studied supports

 

3.2. Effect of the addition of Ni

Characterization results by the BET method, isoelectric points, zero point change (ZPC) and fraction of covered surface by metals (XM) by Electrophoretic migration measurements, of Ni (X%) supported on y-Al2O3 adsorbent are shown in Table 2. First, we observed a very slight decrease in the specific area increasing metal content, at higher metal content studied (6% Ni) the decreasing is close to 2% of the initial area of the support, so in reality there isn’t a decrease in specific area with the metal content; probably because the metal is well dispersed on the support surface. Fig. 2 shows measurement of zeta potential (ZP) with the pH of the solution samples y-Al2O3, NiO, Ni(x%)/y-Al2O3, to determine the isoelectric points (IEP) of y-Al2O3 and NiO, and the zero point charge (ZPC) of the adsorbents Ni(x%)/y-Al2O3, which are shown in Table 2.

 

Fig. 2 Electrophoretic migration measurement of Ni(x)/y-Al2O3 adsorbents

 

Table 2: Isoelectric point (IEP), zero point of charge (ZPC),
Fraction of covered surface by metals (XM), Specific areas
measured by BET and adsorption capacities of Ni (X%)
supported on y-Al2O3.

(*) Taken from reference 28

 

In Table 2 we can observe that by impregnating alumina with Ni decreases the ZPC with the metallic content from the IEP of the alumina (8.7) to the ZPC of the adsorbent 4%Ni/y-Al2O3 (6.8); this decrease is due to the formation of species with lower IEP; in this case the formation of aluminate, which has an IEP of 2.7. Until content near to 4%, the Ni would not form NiO species because then the ZPC should be greater than the IEP of alumina. It is also noted that XM increased by forming aluminate dispersed on the alumina surface, reached a value of 0.33 to 4% of Ni.

At higher contents (6% Ni), the ZPC increases from the value of 6.8 (4%) to 7.5 for 6%Ni/y-Al2O3, due to a higher content begins to form NiO (IEP 9.4) on the aluminate; for this adsorbent is not possible to calculate the XM because we have 3 species present28.

In Table 2, finally the values for pyridine adsorption capacity are shown, expressed in mg of pyridine adsorbed per gram of adsorbent, obtained from the area under the adsorption curve. It’s clearly shown that the adsorption capacity decreases with the metal content; from 22.1 mg g-1 (for y-Al2O3) to 14.6 mg g-1, when the nickel content is 4%, the adsorption capacity then increases to 16.1 mg g-1, when the content reaches 6%.

For this results, we see that there is a direct correlation between the presences of NiO on the surface and adsorption capacity, also the presence of aluminate decreases the adsorption capacity of the material.

Furthermore, there is a relationship between the ZCP of adsorbent with adsorption capacity, this relationship is observed better in Fig. 3.

 

Fig. 3 Adsorption capacity of Pyridine (bars) and ZPC (dots)
of Ni supported on alumina adsorbents

 

In Fig. 3, if we observe the ZPC of adsorbents and IEP of alumina; there is a direct relationship between the ZPC and the adsorption capacity; when the ZPC of adsorbent decreases, the adsorbent loses surface positive charge, for example to 4%Ni/y-Al2O3 adsorbent, the pH of feed (near to 7.0) is slightly higher than the ZPC of material (6.8), losing in this case the property of acid Lewis that the adsorbent will possessed at less content or even the support and losing adsorption capacity due to the Lewis base pyridine property.

For 6%Ni/y-Al2O3 adsorbent, the adsorption capacity increases again; because also increases the ZPC to 7.5; due to the presence of NiO surface, while the difference between the pH of the feed and the ZPC of adsorbent is minimal in this case we can also have the factor of adsorption via π-complexation of Ni(II).

Additionally, we have carried out the adsorption of quinoline with the same adsorbents used in this work, it’s interesting that despite the molecular size of the quinoline, better adsorption capacities are observed for the three adsorbents, this allows us to deduce that adsorption by π-complexation could be used to the treatment of refractory molecules, this results will be analyzed in a future work.

4. CONCLUSIONS

The study of the adsorption capacity for the three different supports shows that not exists a direct relationship between the BET area of the support and its adsorption capacity, however and important factor correspond to the IEP of the support, being y-Al2O3 who possesses the highest adsorption capacity, this is because y-Al2O3 have a positive net charge on his surface favoring the adsorption of pyridine which acts as a Lewis base. The incorporation of Ni on the support generates a decrease in the adsorption capacity of the support, which depends on the ZPC, associated with the presence of aluminates and NiO on the surface of the support.

ACKNOWLEDGMENT

The authors are grateful to the Chilean government for financial support from CONICYT through FONDECYT grant 1140808.

5. REFERENCES

1. J.B. Speight, Chemistry and Technology of Petroleum, New York (1998)        [ Links ]

2. S. Shin, K. Sakanishi, I. Mochida, D.A. Grudoski, J.H. Shinn. Energy & Fuels 14, 539-544 (2000)        [ Links ]

3. G.C. Laredo, S. Leyva, R. Alvarez, M.T. Mares, J. Castillo, J.L. Cano, Fuel 81, 1341-1350 (2002)        [ Links ]

4. P.Grange, Catal. Rev.-Sci.Eng. 21, 135-181 (1980).         [ Links ]

5. A.K. Mathur, C.B. Majumder, S. Chatterjee, J. Hazard. Mater. 157, 335-343 (2008)        [ Links ]

6. K.V. Padoley, A.S. Rajvaidya, T.V. Subbarao, Biores. Technol. 97, 1225-1236 (2006)        [ Links ]

7. L. Qiao, J. Wang, J. Hazard. Mater. 176, 220-225 (2010)        [ Links ]

8. H. Zhao, S. Xu, J. Zhong, X. Bao, Catal. Today 93-95, 857-861 (2004).         [ Links ]

9. R. Andreozzi, A. Insola, V. Caprio, M.G. D’Amore, Water Res. 25, 655-659 (1991).

10. M.K. Mandal, P.K. Bhattacharya, J. Membrane Sci. 286, 115-124 (2006)        [ Links ]

11. M.E. Essington, Environ. Geol. Water Sci. 19, 83-89 (1992).         [ Links ]

12. D.H. Lataye, I.M. Mishra, I.D. Mall, J. Hazard. Mater. 154, 858-870 (2008)        [ Links ]

13. D. Mohan, K.P. Singh, S. Sinha, D. Gosh, Carbon 43, 1680-1693 (2005)        [ Links ]

14. J. Ren, J. Wang, C. Huo, X. Wen, Z. Cao, S. Yuan, Surf. Sci. 601, 1599-1607 (2007)        [ Links ]

15. P. Baeza, G. Aguila, F. Gracia, P. Araya, Catal. Commun. 9, 751-755 (2008).         [ Links ]

16. P. Baeza, G. Aguila, G. Vargas, J. Ojeda, P. Araya, Appl. Catal. B 111, 133-140 (2012)        [ Links ]

17. A.J.Hernandez-Maldonado, R.T. Yang, AIChE Journal 50(4), 791-801 (2004)        [ Links ]

18. F. Aparicio, E. Camú, M. Villarroel, N. Escalona, P. Baeza, J. Chil. Chem. Soc. 58, 2057-2060 (2013).         [ Links ]

19. O. Hamdaoui, E. Naffrechoux. J. Hazard Mater. 147, 381-394 (2007)        [ Links ]

20. A.K. Bajpai, M. Rajpoot, D.D. Mishra, J. Colloid Interface Sci. 187, 96-104 (1997)        [ Links ]

21. C. Huang, J. Colloid Interface Sci. 53, 178-186 (1975)        [ Links ]

22. E. Baumgarten, U. Kirchhausen-Dusing, J. Colloid Interface Sci. 194, 1-9 (1997)        [ Links ]

23. T. Tranior, G. Brown Jr., G. Parks, J. Colloid Interface Sci. 231, 359-372 (2000)        [ Links ]

24. J. Kwon, J. Moon, Y. Bae, D. Lee, H. Sohn, C. Lee. ChemSusChem. 1, 307-309 (2008)        [ Links ]

25. A. Koriakin, K. Muruganandam, C. Lee. Chem. Eng. J. 162, 649-655 (2010)        [ Links ]

26. J.-P. Bellat, I. Bezverkhyy, G. Weber, S. Royer, R. Averlant, J.-M. Giraudon and J.-F. Lamonier, J. Hazard. Mater. 300 711-717 (2015)        [ Links ]

27. H. Yamada, D. S. Dao, J. Fujiki and K. Yogo, Sep. Sci. Technol. 50, 2948-2953 (2015) Department of Chemical and Biomolecular Engineering, Yonsei University, 262 SeongSanno, Seodaemun-gu, Seoul 120-749, South Korea Received 6 April 2010, Revised 9 June 2010, Accepted 9 June 2010, Available online 17 June 2010        [ Links ]

28. F.J. Gil Llambias, A.M. Escudey, J. Santos-Blanco, J. Catal. 83, 225-228 (1983).         [ Links ]

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License