- Email: [email protected]
Extraction and recovery of toxic acidic components from highly acidic oil using ionic liquids
Fuel 181 (2016) 579–586
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Extraction and recovery of toxic acidic components from highly acidic oil using ionic liquids Syed Nasir Shah a, Mohd Ismail e, Mohammad Ibrahim Abdul Mutalib b, Rashidah Binti Mohd Pilus d, Lethesh Kallidanthiyil Chellappan a,c,⇑ a
Centre of Research in Ionic Liquids, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia Department of Chemical Engineering, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia Center for Biofuel and Biochemical Research, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia d Department of Petroleum Engineering, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia e Department of Chemistry, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia b c
h i g h l i g h t s Highly toxic naphthenic acid was extracted from highly acidic model oil using ionic liquids. The extraction process required only very low amount of ionic liquids. Extracted naphthenic acid and the ionic liquids used were recovered by simple procedure. DFT calculations were performed to understand the mechanism of naphthenic acid extraction using ionic liquids. The ionic liquids can be used more than six times without losing its extraction efficiency.
a r t i c l e
i n f o
Article history: Received 22 March 2016 Received in revised form 6 May 2016 Accepted 7 May 2016
Keywords: Ionic liquids Naphthenic acid Thiocyanate anion Regeneration Extraction
a b s t r a c t Naphthenic acid (NA) is a toxic compound that exists in the effluent discharged from highly acidic oil refineries. The amount of NA present in acidic crude oil can be as high as 4 wt%. The complicated structure of NA poses a challenge for oil refineries in their effort to extract NA from the heavy crude oil in an economical and environmental friendly manner. In the current study the extraction of NA from highly acidic model oil by ionic liquid (ILs) was performed using 1,8-diazobicycloundec-7-ene (DBU) based cation in combination with the thiocyanate anion. A detailed computer simulation study on the mechanism of NA extraction by the ILs was also performed. The extracted NA was completely recovered and the ILs used were regenerated by simple addition of water. It was found that increasing of the alkyl chain length increases the percent NA removal. Computer simulation suggests those thiocyanate anions are found to be playing a major role in the NA extraction process. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction The increase in the processing of acidic crude by refineries had raised significant environmental and operational concerns [1,2]. The major problem associated with the acidic crude oil is the presence of NA, which is considered as both toxic and corrosive. NA is generally described by the general formula CnH2n+zO2, where n indicates the number of carbon atoms; z (negative integer or zero) indicates the deficiency of hydrogen because of the presence of cyclic or aromatic groups [3–5]. NA is highly toxic to aquatic organ⇑ Corresponding author at: Center for Biofuel and Biochemical Research, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia. E-mail address: [email protected] (L.K. Chellappan). http://dx.doi.org/10.1016/j.fuel.2016.05.041 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
isms and is mostly present in the effluent discharge from different petrochemical industries processing acidic crude oil. Most of the studies focus on the NA toxicity in the effluent discharge. In order to avoid the toxic effect of NA acid on the aquatic system, a simple and efficient process for the extraction of NA from its source of origin (acidic crude oil) would be beneficial [6–8]. NA have higher solubility in water compared to hydrocarbons thus making it a major threat to aquatic environment [9]. The presence of NA can be as high as 4 wt% in heavy crude oil. The NA can also cause corrosion in the oil refineries and it can adversely affect the storage and combustion properties of the final product. On the other hand NA is an important raw material (2500–3500 USD/ton) having potential application in different areas such as production of paint and cross linking of rubber [10]. Hence, the removal and recovery of NA are
580
S.N. Shah et al. / Fuel 181 (2016) 579–586
highly advantageous to both the environment and to the industry. The presence of NA in crude oil is measured using the total acid number (TAN). TAN values greater than 0.5 mg KOH/g of crude oil are considered as potentially corrosive for refineries units [11]. The current method used in the industry for the extraction and recovery of NA is the soda wash method, which leads to the spillage of NA in the water reservoirs [10]. Adsorption using ion exchange and liquid–liquid extraction (LLE) using organic solvents are the methods that have been tested for the recovery of NA from crude oil at lab scale [12–14]. Decarboxylation is another alternative for acidity reduction of highly acidic crudes [15,16]. Esterification using SnO/Al2O3 had been used for diesel streams as well as for crude oil [17,18]. However, all these methods have certain operational difficulties such as harsh reaction conditions, higher operational costs and lack of environmental viability. The restrictive factors in the current methods demands for a simple, environmental friendly and industrially viable process for the extraction of NA. Ionic liquids (ILs), solvents composed entirely of ions, have found application in energy, nuclear, catalysis, biomass, electrochemical and solar cells industries [19–26]. Although different ILs are used for extraction of NA, most of them are applicable only in the case of very low total acid number (<0.5 TAN). Moreover, most of them are using large amount of volatile organic solvents and the regeneration procedure of ILs and NA is quite complicated [27–31]. It has been shown that using anion with high pKa can increase the extraction efficiency of NA [28,30]. This essentially suggests that the more basic the IL is, the higher extraction efficiency will become. However the effect of the basicity of the cation has not been tested. It has also been proposed that the mechanism of NA extraction using ILs occur through the formation of a cage-like structure of the IL species around the NA molecules [30]. However, no details study about the mechanism has been performed to date. In this work, the extraction of NA using diazabicyclo undecane (DBU) based cation and thiocyanate anion is studied. The DBU based cation was selected for the extraction studies because conjugate acid of DBU is more basic than the conjugate acid of imidazole (>11 vs 7) [34,35]. In theory the DBU-based cation should have higher extraction efficiency. The thiocyanate anion is incorporated because of the presence of two coordination site, which will lead to the enhanced interaction between ILs and NA. In addition, computational study is performed to understand the molecular level mechanism of the extraction process. 2. Experimental 2.1. General information The chemicals used in this study were purchased from Sigma– Aldrich (Bornem, Belgium) and Acros Organics (Geel, Belgium). The synthesis and characterization of the ILs used in this study are performed using previously reported procedure [32]. The ILs used in this study are 1-hexyl-1,8-diazobicycloundec-7-ene thiocyanate, [DBU] [Hex], 1-octyl-1,8-diazobicycloundec-7-ene thiocyanate, [DBU] [Oct], 1-decyl-1,8-diazobicycloundec-7-ene thiocyanate, [DBU] [Dec]. The NMR spectra are recorded using Bruker Advance 500 spectrometer (operating at 500 MHz for 1H and 125 MHz for 13C). Number of scans is 16 and the pulse programme used is zg30. 2.2. Preparation of model acid oil In the current study highly acidic oil is used for de-acidification. The total acid number of the acidic oil used in this study is 3.46 (±0.01) mg KOH/g of oil, which is calculated using Metler Toledo
Auto titrator according to the ASTM D664 method. The acidic oil is prepared by adding commercial NA to dodecane. Dodecane is used in this study because of its closed proximity with kerosene, jet fuel oil and diesel. All these streams have the maximum concentration of naphthenic acid [33]. 2.3. Deacidification process The de-acidification of acidic oil is done by mixing model oil and ILs in a round bottom flask at constant stirring for one hour. The temperature is adjusted using silicon bath controlled by a hot plate and thermocouple. Once the reaction has completed, the reactants are transferred into a separating funnel. The two layers were easily separated because of the density difference. The de-acidified model oil is collected from the top of the separating funnel and its total acid number is analyzed. The percent NA is calculated using the formula given below:
Percent Naphthenic Acid Remov al ¼
1
TANf TANi
100;
where TANf and TANi refer to final and initial TAN of the oil respectively. 2.4. Computational details In order to understand the mechanism of the naphthenic acid extraction with DBU/SCN-based ILs, molecular dynamics simulation was performed on the [DBU-Hex] [SCN] system. To obtain reasonable statistics, the system is comprised of 1527 n-dodecane molecules, 86 [DBU-Hex] [SCN] ILs molecules, and 10 cyclohexyl acetic acid as the model NA. The simulation is performed using the AMBER [34] molecular dynamics package employing the GAFF-force field [35]. The charges were obtained using the RED Server Tools [36–39]. One modification was made to the angular force constant of the SACAN. The modification was made due to the unmodified GAFF force constant causing the SCN molecule to be mostly in the bent form, while from our geometry optimization with ab initio method suggests that the SCN molecule is in linear form at equilibrium, and thus should be in the linear form most of the time. A molecular dynamics simulation on the pure [DBUHex] [SCN] system gives a density value of 1.0419 g/mL, which is in good agreement with the experimental density of 1.0564 g/mL obtained in our lab. The percent error is calculated to be 1.3726%. The initial configuration of the system namely the topology as well as the input coordinate files required for the simulation are obtained using Avogadro [40], Packmol [41] and AmberTools 14 [42] software packages. The molecules are fitted into a cubic box using Packmol. The initial configuration of the system is minimized using the steepest decent method for 30,000 steps. Next, the system is equilibrated in the constant pressure-constant temperature ensemble (NPT) using Langevin dynamics with a time step of 0.5 fs, and collision frequency of 1.0 ps1. The temperature of the system is set to 300 K while the heat bath temperature coupling was set to 1.0 psi. The pressure of the system is set to 1 bar with isotropic position scaling while the pressure relaxation time was set to 1.0 psi. Periodic boundary condition is used to account for the boundary effect. The Van der Waals cutoff and the direct space-inverted space Particle-Mesh Ewald cutoff are set to 10 Å. The equilibration step was performed for 150 ns. The system is set to have reached equilibrium by making sure that the density and the potential energy of the system were stable. The plots of the density and the potential energy of the system are shown in Figs. 1 and 2 in the supporting information.
581
S.N. Shah et al. / Fuel 181 (2016) 579–586
3. Results and discussions 3.1. Effect of ionic liquid/oil ratio The most important parameter in the extraction study is the ILs to oil ratio. In this study three different ILs having different alkyl chain length are used. The ILs used in this study are shown in Fig. 1. The effect of different IL/oil ratio is investigated by varying the amount of ILs. The effect of IL/oil ratio is given in Table 1 and Fig. 2. The increase in the IL/model oil ratio increased the percent of NA extracted for all ILs used in the study. The extraction efficiency of the ILs follows the order [DBU-Dec] [SCN] > [DBU-Oct] [SCN] > [DBU-Hex] [SCN]. Longer the alkyl chain length of ILs, greater the extraction efficiency. When the extraction is performed at room temperature, [DBU-Dec] [SCN] is found to be the most effective in NA extraction. [DBU-Dec] [SCN] extracted about 35% NA at an IL/model oil ratio of 0.010, while [DBU-Oct] [SCN] extracted 32% NA. The lowest extraction performance is observed for [DBUHex] [SCN] (8.83%). The [DBU-Oct] [SCN] and [DBU-Dec] [SCN] showed similar extraction performance until IL/oil ratio of 0.03. When IL/oil ratio exceeded 0.03, the extraction performance between [DBU-Oct] [SCN] and [DBU-Dec] [SCN] increased significantly but showed different values. For instance, [DBU-Dec] [SCN] showed 100% NA extraction at an IL/model Oil of 0.06, while [DBU-Oct] [SCN] showed only 91% extraction performance at similar IL/model oil ratio. The reason for this difference could be due to the increased Van der Waals interaction between the NA and the longer alkyl chain in [DBU-Dec] [SCN] and the higher miscibility of n-dodecane with [DBU-Dec] [SCN] [31]. Among the ILs studied, [DBU-Hex] [SCN] showed the lowest extraction performance. For example, [DBU-Hex] [SCN] showed 100% extraction performance at an IL/model oil ratio of 0.10. As reported earlier [DBU-Oct] [SCN] and [DBU-Dec] [SCN] showed 100% extraction efficiency at a lower IL/model Oil ratio. The lower extraction performance of [DBU-Hex] [SCN] is related to its shorter chain length and higher viscosity [43]. The shorter alkyl chain length reduces the extent of Van der Waals interaction with the NA and the higher viscosity prevents the effective mass transport of the NA moiety. The imidazolate anion based ILs reported by Yu Sun et al. with [C8mim]+ cation showed 100% extraction efficiency at an IL/model
Table 1 Effect of the IL/oil ratio on naphthenic acid removal. IL/oil ratio (wt%)
0.010 0.020 0.030 0.040 0.050 0.06 0.07 0.08 0.09 0.1
100
Percent Naphthenic Acid Removal
After the system has attained equilibrium, the production run is performed in the NPT ensemble for 10 ns, using similar parameters as for the equilibration run. The trajectory of the 10 ns production is analyzed to study the mechanism of the naphthenic acid extraction by the ionic liquid.
Percent NA removal [DBU-Hex] [SCN]
[DBU-Oct] [SCN]
[DBU-Dec] [SCN]
8.83 21.32 35.09 46.98 57.17 66.12 75.24 82.99 91.76 100
32.39 48.74 63.76 76.26 85.61 91.31 99.85 100 100 100
34.93 52.51 67.86 82.26 93.80 100 100 100 100 100
[DBU-Hex][SCN] [DBU-Oct][SCN] [DBU-Dec][SCN]
80
60
40
20
0 0.02
0.04
0.06
0.08
0.10
IL/Oil ratio (W/W) Fig. 2. Effect of IL/oil ratio on percent naphthenic acid removal (j, [DBU-Hex] [SCN]; d, [DBU-Oct] [SCN]; N, [DBU-Dec] [SCN]) (stirring rate = 500 RPM, reaction time = 1 h).
oil ratio of 0.008 with a total acid number of 0.20 [30]. This extremely low total acid number do not even satisfied the criteria of acidic crude oil, where the total acid number should be greater than 0.5 mg KOH/g of oil [33]. In one of other studies [C8mim]+ cation with phenolate anion showed 100% extraction efficiency at an IL/oil ratio of 0.09 and total acid number of the model oil is 1.44 mg KOH/g of oil [27]. In comparison to the work reported by Yu Sun et al. [DBU-Oct] [SCN] requires higher IL/model ratio (0.07) for complete de-acidification of the model oil. One of the reasons for this relatively high IL/oil ratio is the high acidic oil used in the current study that having a total acid number of 3.46 mg KOH/g of oil. The thiocyanate based ILs perform exceptionally well as compared to N-alkyl pyridinium bromides and 1-alky-3methylimidazolium bromides [31]. For instance, 1-octyl-3methylimidazolium bromide, [C8mim] [Br], extracted only 93% of NA from a model oil with an IL/model oil ratio of 0.1, even if the acidity of the model system was very low (<0.5 TAN). At the similar conditions, 1-octylpyridinium bromide extracted less than 65% of NA. 3.2. Effect of temperature on naphthenic acid extraction
Fig. 1. Overview of ionic liquids used in this study.
The effect of temperature on the de-acidification is investigated using [DBU-Dec] [SCN] in a temperature range of 303.15–383.15 K. [DBU-Dec] [SCN] is used because of its higher extraction efficiency. Other extraction parameters are kept constant and the results are shown in Fig. 3. It is found that temperature do not have a significant influence on the extraction efficiency of ILs.
582
S.N. Shah et al. / Fuel 181 (2016) 579–586 100
Percent Naphthenic Acid Removal
Percent Naphthenic Acid Removal
72 71 70 69 68 67 66 65
90
80
70
60
50 300
320
340
360
1
380
2
3
4
Temperature
Only three percent enhancement was observed when the temperature was increased from 303.15 K to 383.15 K. The extraction performance slightly increased when the temperature was increased from 303.15 K to 360.15 K. However, there was no increase in extraction efficiency when the temperature increased above 360.15 K. The small increment in the extraction performance with the increase in temperature can be attributed to the enhanced mass transport due to the lowering of the viscosity of the ILs at high temperature. The 1H NMR spectrum of the ILs was recorded after the extraction process in order to confirm the stability of the ILs under elevated temperature. The 1H NMR spectra showed no sign of decomposition of the ILs used (Fig. 3 supporting information). 3.3. Recyclability of ionic liquids The recyclability of the materials used in a process is the most important parameter determining its sustainability and industrial viability. The extent of reusability of the ILs would be a reflection of the sustainability of the current process. The reusability of ILs is investigated using [DBU-Dec] [SCN]. An IL/oil ratio of 0.05 is used and the extraction is performed at room temperature. A model oil system having a TAN strength of 0.6 mg KOH/g of oil is used. The model oil is completely separated from the ILs after each extraction and the percentage of NA removal is calculated by measuring the TAN. The procedure is repeated by using fresh model oil and the previously used ILs. The experiment is repeated until the NA extraction efficiency of the ILs is reduced to less than 60% and the results are shown in Fig. 4. It is clear from the experimental results that the DBU based ILs can be recycled more than six times without significant reduction in their extraction efficiency. The extraction efficiency remained more than 95% even after seven cycles of extraction and DBU derived ILs showed more than 50% extraction efficiency even after they have been used more than 10 times. 3.4. Effect of different total acid number of oil The different TAN strengths of the model oil can also affect the extraction efficiency. The efficiency of the DBU derived ILs for the deacidification of model oil with different TAN strength is investigated by preparing model oil system with TAN up to 8 mg KOH/g. The study is performed using [DBU-Dec] [SCN] with the optimized IL/oil ratio of 0.05 and the results are shown in Fig. 5. The TAN
6
7
8
9
10
Fig. 4. Recyclability study of ionic liquids (j, [DBU-Dec] [SCN] with 0.05 IL/oil ratio at 303 K) (stirring rate = 500 RPM, reaction time = 1 h).
100
Percent Naphthenic Acid Removal
Fig. 3. Effect of temperature on naphthenic acid removal (j, [DBU-Dec] [SCN] with 0.03 IL/oil ratio) (stirring rate = 500 RPM, reaction time = 1 h).
5
No of Reactions
90 80 70 60 50 40 30 20
1
2
3
4
5
6
7
8
Total Acid Number Fig. 5. Effect of increase in total acid number on extraction efficiency of naphthenic acid (j, [DBU-Dec] [SCN] with IL/oil ratio of 0.05; (stirring rate = 500 RPM, reaction time = 1 h).
strength of the model oil is increased by adding more NA to the model oil. The extraction and separation time is set at one hour. The [DBUDec] [SCN] is able to completely de-acidify the model oil with a TAN value of 3.0 mg KOH/g. Although the TAN value of the model oil is increased significantly (8 mg KOH/g), the DBU derived ILs showed a deacidification efficiency of more than 40%, which indicates their potential in deacidification of highly acidic model systems.
3.5. Recovery of naphthenic acid Besides de-acidifying crude oil, the recovery of NA after the extraction with ILs is important due to its potential utilization in the chemical industry. Hence, a byproduct could be produced and it will lead to better economic potential for the proposed extraction process. The recovery of the NA from the ILs was studied for the purpose of exploring the extent of possible NA to be recovered as a value added product. The NA is recovered by adding deionized water into the ILs layer. Two layers were formed because of the hydrophobic nature of the NA and the hydrophilic nature of the ILs used ([DBU-Hex]
583
S.N. Shah et al. / Fuel 181 (2016) 579–586
[SCN]). The NA is separated from the IL layer in water. After this, NA is dried under vacuum and subjected to FTIR (Fig. 4 in supporting information) and 1H NMR analysis (Fig. 5 in supporting information). Both 1H NMR and FTIR spectrum was found to be similar to that of the pure NA. The FTIR spectrum of the pure NA showed the presence of several important functional groups that characterize it. The peak at 1698 cm1 confirmed the presence of carbonyl group (AC@O) whilst the strong peak at 1258 cm1 represents the CAO stretch in carboxylic acid. The broad peak at 2956 cm1 corresponds to the CAH stretch and another peak at 1463 cm1 is due to the CAH bend in alkanes. More than 97% of the NA appeared in the model oil was recovered by the addition of water. The ILs used for NA extraction was regenerated by the evaporation of the water using a rotary evaporator and the 1H NMR spectra of the recovered ILs was similar as in the case of pure ILs (Fig. 6 in supporting information). 3.6. Molecular dynamics simulation of ionic liquid, naphthenic acid, and model oil There were 3 possible mechanisms by which the NA could be removed by ILs namely: (i) the extraction of the acid by the anionic species; (ii) the formation of liquid clathrate in the sustaining medium with cage-like structure of the ILs surrounding the NA; (iii) the p–p interaction between the NA and the cationic species [30]. The molecular dynamics simulation performed in this study is based on classical mechanics nature and thus will not reveal any p–p interaction phenomenon between the NA and the DBU-based cation, which is of electronic nature. Nevertheless, the molecular dynamics simulation should be able to model the extraction of NA by the anionic species, and the formation of cage-like structure of the ionic liquids around the naphthenic acid. To look for the cage-like structure, the relative position of the DBU-cation, thiocyanate anion, and the model oil around the NA molecule as their trajectory is integrated over time is determined.
To simplify the analysis, N1-nitrogen atom of the DBU-cation, the C-carbon atom of the thiocyanate anion, and the C, C5, C6, and C11-carbon atom of the n-dodecane model oil are used as the reference as shown in Fig. 6. Fig. 7 shows the radial distribution function (RDF) plot of all the ILs and model oil molecules around a representative NA. Based on the plot, it can be deduced that the first shell surrounding the NA can be complicated with three different species surrounding the NA, as evidenced by the two extra shoulders on the first peak. Fig. 8 shows the radial distribution function (RDF) plot of the DBU-cation, the SCN anion, and the n-dodecane model oil from the geometrical center of the cyclohexyl acetic acid molecule. The radial distribution function (RDF) plot of all the 10 types of NA showed that the average distance from the center of geometry for the NA for the N1-nitrogen on DBU-cation is about 7.1 Å, While the average distance to the nearest thiocyanate anion (carbon atom) is on average 5.4 Å and average distance to the nearest ndodecane model oil is 5.7 Å. The results suggest that the NA has higher tendency to be surrounded by the model oil molecules first, followed by thiocyanate anions and the least is DBU cations since it is the furthest from the center of geometry of the NA. This might suggest that the DBU-cation is just a spectator ion attracted to the thiocyanate anion and not adjacent to the NA, thus forming a cage-like structure as suggested. In addition, the model oil is also very close to the NA molecule, suggesting that if there is any cage-like structure formed within the system, it also includes the model oil in it. A better conclusion is that there is no cage-like structure formed by the cation and the anion around the NA as suggested previously. From the integration of the first RDF peaks, the number of DBUcation surrounding the naphthenic acid molecule varies between 1 and 3, whilst for the thiocyanate anion it is between 1 and 2, and for the n-dodecane model oil surrounding the NA it is between 2 and 5. The reasonably high number of DBU-cation suggests there is a possibility that the cation could form a cage-like structure
C5 C
C11 C6
OH S
C5
N
C
O
carbonyl
N
cyclohexyl N1
N
Fig. 6. The position of the atoms used in the simulation.
584
S.N. Shah et al. / Fuel 181 (2016) 579–586
36
1.2
32
1.0
DBU-cation SCN-anion Dodecane
28
0.8
24 20
g (r)
g (r)
0.6
0.4
16 12 8
0.2
4
0.0 0
0
3
6
9
12
15
18
0
21
5
10
15
20
Distance (Angstrom)
Distance (Angstrom) Fig. 7. The RDF of ionic liquid and the model oil around a representative naphthenic acid.
Fig. 9. The RDF of the DBU-cation, SCN-anion, and the n-dodecane model oil from the center of geometry of the carbonyl moiety of the naphthenic acid.
12
16
DBU-cation SCN-anion Dodecane
14 12
DBU-cation SCN-anion Dodecane
10
8
g (r)
g (r)
10 8 6
6
4
4 2
2 0
0 0
4
8
12
16
20
Distance (Angstrom)
0
4
8
12
16
20
Distance (Angstrom)
Fig. 8. The RDF of each of the DBU-cation, SCN-anion, and n-dodecane around a representative naphthenic acid.
Fig. 10. The RDF of the DBU-cation, SCN-anion, and the n-dodecane model oil from the center of geometry of the C5-cyclohexyl moiety of the naphthenic acid.
around the NA. However, the low number of thiocyanate anion, and the high number of n-dodecane model oil supports the notion that the cage like structure does not exist. Although the DBU-cation molecules are close enough to form cages surrounding the NA molecules, the distribution of the DBUcation is mostly near the polar side of the NA. This is evidenced from Figs. 9 and 10. Fig. 7 depicts the RDF plot of the center of geometry for the carbonyl moiety (CAOAO) to the N1-nitrogen of the DBU-cation, to the C-carbon of the thiocyanate anion, and to the terminal and inner carbon (C, C5, C6, C11) of the n-dodecane model oil. The average distance from the center of geometry of the carbonyl moiety to the DBU-cation is 5.1 Å, while the same average distances to the C-carbon of the thiocyanate anion and to the n-dodecane model oil are 4.8 and 4.9 Å respectively. Fig. 10 shows the RDF of the C5-carbon of the cyclohexyl ring moiety of the NA to the N1-nitrogen of the DBU-cation, to the C-carbon of the thiocyanate anion, and to the terminal and inner carbon (C, C5, C6, C11) of the n-dodecane molecule. The average distance from the C5-carbon to the N1-nitrogen of the DBU-cation is 8.8 Å, while the average distance to the C-carbon of the thiocyanate anion is 7.4 Å, and the average distance to the n-dodecane model oil is 4.7 Å. From these two figures, it is evidenced that only the carbonyl part of the NA is likely to be surrounded by the DBU-cation and the
thiocyanate anion, while the cyclo hexyl part of the NA is only surrounded by the n-dodecane model oil. Fig. 11 shows a snapshot of one of the NA molecules from the simulation. From these figures, it is evidenced that while the DBU-cation and the thiocyanate anion do surround the NA, they are only located on the polar side of the NA (i.e. the carbonyl moiety), whereas the cyclohexyl ring is mostly surrounded by the n-dodecane model oil. These evidences do not support the idea of the cage-like structure formation around the NA. Based on the presented argument above, the most likely interaction between the ILs and the NA is through the hydrogen bonding between the OH–hydrogen of the NA to the anion of the ILs. To confirm the interaction mechanism involving the OH–hydrogen of the acid, the RDF plot of the OH–hydrogen to the sulfur and nitrogen atoms on the thiocynate anion is generated. Furthermore, to investigate probable interaction of the OH–hydrogen with the nitrogen atoms on the DBU-cation, the RDF plot of the nitrogen atoms on the DBU-cation around the OH–hydrogen is also generated. As evidenced from the RDF plot in Fig. 12, the average distance from the OH–hydrogen to the nitrogen atom and the sulfur atom on the thiocyanate anion is 1.77 and 1.97 Å respectively, while the distance from the OH–hydrogen to the N1-nitrogen and N-nitrogen on the DBU-cation is 4.66 and 4.76 Å respectively.
S.N. Shah et al. / Fuel 181 (2016) 579–586
585
thiocyanate anion at any one time. The coordination to one thiocyanate anion at both the sulfur and the nitrogen at the same time is not observed, although this might be because of the low number of NA to thiocyanate anion ratio used in the simulation that makes the probability of the OH–hydrogen to attach to another thiocyanate anion is higher compared to sharing the thiocyanate anion by two OH–hydrogen atoms. 3.7. Mechanism of naphthenic acid extraction
Fig. 11. A snapshot of the naphthenic acid (pink) is being surrounded by the DBUcation (green), SCN-anion (white), and the n-dodecane model oil (gray). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
80 70
SCN-Sulfur SCN-Nitrogen DBU-N-Nitrogen DBU-N1-Nitrogen
60
g (r)
50 40 30 20 10 0 0
4
8
12
16
20
Distance (Angstrom) Fig. 12. The RDF of the thiocyanate sulfur, thiocyanate nitrogen, DBU-N-nitrogen, and DBU-N1-nitrogen around the OH–hydrogen.
From the experimental results, the ILs with DBU-cation and thiocyanate anion displayed better performance than other previously published ILs used for extraction of NA. It was claimed that the pKa of the anion plays an important role in NA extraction [27,30]. However, from our results, it is evidenced that the pKa is not of significance. A comparison between the use of phenolate anion which has a pKa value of 10 [27], imidazole-based anion used in the previous study [30] had a pKa value higher than 10. However, the pKa of thiocyanate anion is around 1 [44]. If the pKa of the anion plays a big role in ILs extraction of NA, the DBU-cationthiocyanate anion ILs would have had a lower NA extraction values. However, as evidenced by our results, the NA extraction is higher with the DBU-cation-thiocyanate anion ILs. In addition to that, the pKa describes the ability of a negatively charge species to abstract protons from a hydrogen donor. In the NA extraction process by ILs, the anion does not need to abstract the proton from the acid. Instead, the anion only needs to coordinate to the NA thus pulling the acid out from the model oil. From our results, we suspect that the number of available interaction site is the main factor causing the high NA extraction. Not only the thiocyanate anion has lower molecular weight thus increasing the number of thiocyanate anion for each gram of the ILs used in the extraction, it also has two coordination sites for each thiocyanate molecule. This doubles the coordination probability by the ILs and thus increasing the number of NA molecule extracted from the model oil during each extraction cycle. It has also been suggested that the cation and the alkyl side chain plays a role in extracting the NA [27,30,31]. On the contrary our simulation results show that the cation does not get close to the NA molecule thus rendering it only as a spectator in the whole process. It can be postulated that the alkyl chain on the cation increases the ability of the polar cation to be dissolved in the non-polar model oil thus increasing the probability of the anion to interact with the NA molecule. More studies are needed to provide the exact reason of the increase in the NA extraction with the increase in the alkyl chain length. 4. Conclusion
Given the much shorter average distance between the OH–hydrogen to the sulfur and nitrogen atoms on the thiocynate anion, it suggests that the NAAOH interacts only with the thiocyanate anion and not the DBU-cation. Hence it can be deduced for all the 10 NA molecules, the OH– hydrogen interacts dominantly with the nitrogen or the sulfur atoms of the thiocyanate anion. The site of the coordination preferred by the OH–hydrogen is the sulfur atom on the thiocyanate, as evidence by the integration of the first RDF peak, with the sulfur being on average 70% of the time and the nitrogen being on average 30% of the time coordinated to the OH–hydrogen. The thiocyanate has two coordination sites to form hydrogen bonding with the OH–hydrogen from the NA. The two coordination sites on the thiocyanate anion is contrary to the other anions reported and this might have provided the advantage for the thiocyanate anion over others used in the ILs extraction of NA. Nevertheless only one NA is attached to one
The thiocyanate based ILs are very effective in the deacidification of high TAN model oil with an extremely low IL/oil ratio. DBU based cation with alkyl groups such as hexyl, octyl, and decyl were used to study the effect of alkyl groups on extraction efficiency. The extraction efficiency of the DBU-based thiocyanate ILs increases with the increase in the alkyl spacer length on the DBU cation. [DBU-Dec] [SCN] showed maximum extraction efficiency. However, simulation study using the DBU-based cation suggests that it is not directly coordinated to the NA in the extraction process. Instead, the thiocyanate anion is the species coordinated directly with the NA. The simulation study showed that thiocyanate anion has two possible coordination sites. This increases the capability of the thiocyanate anion to interact with the NA from the oil, thus increasing the ability of the thiocyanate-based ILs to abstract NA. All the ILs can be reused more than six times for the extraction without compromising their
586
S.N. Shah et al. / Fuel 181 (2016) 579–586
extraction performance. More than 98% of the extracted NA was regenerated by the simple addition of water into the ILs. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fuel.2016.05.041. References [1] http://www.apecconsulting.com/PDF/GlobalHighTAN.pdf. [2] Pedro P, Roger M, Luis Z. AquaconversionTM: a new option for residue conversion and heavy oil upgrading. J Vis Technol 1998;6:5–13. [3] Fan TP. Characterization of naphthenic acids in petroleum by fast atom bombardment mass spectrometry. Energy Fuel 1991;5:371–5. [4] Hsu CS, Dechert G, Robbins W, Fukuda E. Naphthenic acids in crude oils characterized by mass spectrometry. Energy Fuel 2000;14:217–23. [5] Rudzinski WE, Oehlers L, Zhang Y, Najera B. Tandem mass spectrometric characterization of commercial naphthenic acids and a Maya crude oil. Energy Fuel 2002;16:1178–85. [6] Thomas K, Langford K, Petersen K, Smith A, Tollefsen K. Effect-directed identification of naphthenic acids as important in vitro xeno-estrogens and anti-androgens in North Sea offshore produced water discharges. Environ Sci Technol 2009;43:8066–71. [7] Giesy JP, Anderson JC, Wiseman SB. Alberta oil sands development. Proc Nat Acad Sci 2010;107:951–2. [8] Frank RA, Fischer K, Kavanagh R, Burnison BK, Arsenault G, Headley JV. Effect of carboxylic acid content on the acute toxicity of oil sands naphthenic acids. Environ Sci Technol 2008;43:266–71. [9] Headley JV, McMartin DW. A review of the occurrence and fate of naphthenic acids in aquatic environments. J Environ Sci Health Part A 2004;39:1989–2010. [10] Brient JA, Wessner PJ, Doyle MN. Naphthenic acids. Kirk-Othmer Encyclopedia of Chemical Technology; 1995. [11] Testing ASf, Materials. ASTM D-664: standard test method for acid number of petroleum products by potentiometric titration. Annual book of ASTM standards; 1989. [12] Holowenko FM, MacKinnon MD, Fedorak PM. Characterization of naphthenic acids in oil sands wastewaters by gas chromatography–mass spectrometry. Water Res 2002;36:2843–55. [13] Gaikar V, Maiti D. Adsorptive recovery of naphthenic acids using ion-exchange resins. React Funct Polym 1996;31:155–64. [14] Hoeiland S, Barth T, Blokhus A, Skauge A. The effect of crude oil acid fractions on wettability as studied by interfacial tension and contact angles. J Pet Sci Eng 2001;30:91–103. [15] Ding L, Rahimi P, Hawkins R, Bhatt S, Shi Y. Naphthenic acid removal from heavy oils on alkaline earth-metal oxides and ZnO catalysts. Appl Catal A 2009;371:121–30. [16] Zhang A, Ma Q, Wang K, Liu X, Shuler P, Tang Y. Naphthenic acid removal from crude oil through catalytic decarboxylation on magnesium oxide. Appl Catal A 2006;303:103–9. [17] Wang YZ, Sun XY, Liu YP, Liu CG. Removal of naphthenic acids from a diesel fuel by esterification. Energy Fuel 2007;21:941–3. [18] Wang Y-z, Li J-y, Sun X-y, Duan H-l, Song C-m, Zhang M. Removal of naphthenic acids from crude oils by fixed-bed catalytic esterification. Fuel 2014;116:723–8. [19] Shah SN, Lethesh KC, Mutalib M Abdul, Pilus M, Binti R. Extraction and recovery of naphthenic acid from acidic oil using supported ionic liquid phases (SILPs). Chem Prod Proc Model 2015;10:221–8.
[20] MacFarlane DR, Tachikawa N, Forsyth M, Pringle JM, Howlett PC, Elliott GD. Energy applications of ionic liquids. Energy Environ Sci 2014;7:232–50. [21] Ullah Z, Bustam MA, Man Z. Biodiesel production from waste cooking oil by acidic ionic liquid as a catalyst. Renew Energy 2015;77:521–6. [22] Ohno H. Electrochemical aspects of ionic liquids. Wiley.com; 2011. [23] Rao C, Venkatesan K, Nagarajan K, Srinivasan T. Dissolution of uranium oxides and electrochemical behavior of U (VI) in task specific ionic liquid. Radiochim Acta 2008;96:403–9. [24] Swatloski RP, Spear SK, Holbrey JD, Rogers RD. Dissolution of cellose with ionic liquids. J Am Chem Soc 2002;124:4974–5. [25] Wu B, Reddy R, Rogers R. Novel ionic liquid thermal storage for solar thermal electric power systems. Sol Eng 2001:445–52. [26] Ping B, Yusof M. Characteristics and properties of fatty acid distillates from palm oil. Oil Palm Bull 2009;59:5–11. [27] Nasir Shah S, Chellappan L Kallidanthiyil, Gonfa G, Mutalib MIA, Pilus RBM, Bustam MA. Extraction of naphthenic acid from highly acidic oil using phenolate based ionic liquids. Chem Eng J 2016;284:487–93. [28] Nasir Shah S, Mutalib MIA, Pilus RBM, Lethesh KC. Extraction of naphthenic acid from highly acidic oil using hydroxide-based ionic liquids. Energy Fuels 2014;29:106–11. [29] Shi LJ, Shen BX, Wang GQ. Removal of naphthenic acids from Beijiang crude oil by forming ionic liquids. Energy Fuels 2008;22:4177–81. [30] Sun Y, Shi L. Basic ionic liquids with imidazole anion: new reagents to remove naphthenic acids from crude oil with high total acid number. Fuel 2012;99:83–7. [31] Duan J, Sun Y, Shi L. Three different types of heterocycle of nitrogen-containing alkaline ionic liquids treatment of acid oil to remove naphthenic acids. Catal Today 2013;212:180–5. [32] Lethesh KC, Shah SN, Mutalib MA. Synthesis, characterization, and thermophysical properties of 1,8-diazobicyclo [5.4. 0] undec-7-ene based thiocyanate ionic liquids. J Chem Eng Data 2014;59:1788–95. [33] Speight JG. Chapter 1 – Naphthenic acids in petroleum. In: Speight JG, editor. High acid crudes. Boston: Gulf Professional Publishing; 2014. p. 1–29. [34] Case D, Darden T, Cheatham III TE, Simmerling C, Wang J, Duke R. AMBER 12. San Francisco: University of California; 2012. p. 1. [35] Wang J, Wolf RM, Caldwell JW, Kollman PA, Case DA. Development and testing of a general amber force field. J Comput Chem 2004;25:1157–74. [36] Vanquelef E, Simon S, Marquant G, Garcia E, Klimerak G, Delepine JC. RED server: a web service for deriving RESP and ESP charges and building force field libraries for new molecules and molecular fragments. Nucl Acids Res 2011;39:W511–7. [37] Dupradeau F-Y, Pigache A, Zaffran T, Savineau C, Lelong R, Grivel N, et al. The RED. Tools: advances in RESP and ESP charge derivation and force field library building. Phys Chem Chem Phys 2010;12:7821–39. [38] Bayly CI, Cieplak P, Cornell W, Kollman PA. A well-behaved electrostatic potential based method using charge restraints for deriving atomic charges: the RESP model. J Phys Chem 1993;97:10269–80. [39] Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon MS, Jensen JH, et al. General atomic and molecular electronic structure system. J Comput Chem 1993;14:1347–63. [40] Hanwell MD, Curtis DE, Lonie DC, Vandermeersch T, Zurek E, Hutchison GR. Avogadro: an advanced semantic chemical editor, visualization, and analysis platform. J Cheminformatics 2012;4:17. [41] Martínez L, Andrade R, Birgin EG, Martínez JM. PACKMOL: a package for building initial configurations for molecular dynamics simulations. J Comput Chem 2009;30:2157–64. [42] Case D, Babin V, Berryman J, Betz R, Cai Q, Cerutti D. Amber 14; 2014. [43] Lethesh KC, Shah Syed Nasir, Mutalib MI Abdul. Synthesis, characterization, and thermophysical properties of 1,8-diazobicyclo[5.4.0]undec-7-ene based thiocyanate ionic liquids. J Chem Eng Data 2014;59:1788–95. [44] Martell A, Smith R. NIST stability constants of metal complexes database 46, 6.0.. Gaithersburg (MD): US Department of Commerce; 2001.
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Extraction and recovery of toxic acidic components from highly acidic oil using ionic liquids Syed Nasir Shah a, Mohd Ismail e, Mohammad Ibrahim Abdul Mutalib b, Rashidah Binti Mohd Pilus d, Lethesh Kallidanthiyil Chellappan a,c,⇑ a
Centre of Research in Ionic Liquids, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia Department of Chemical Engineering, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia Center for Biofuel and Biochemical Research, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia d Department of Petroleum Engineering, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia e Department of Chemistry, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia b c
h i g h l i g h t s Highly toxic naphthenic acid was extracted from highly acidic model oil using ionic liquids. The extraction process required only very low amount of ionic liquids. Extracted naphthenic acid and the ionic liquids used were recovered by simple procedure. DFT calculations were performed to understand the mechanism of naphthenic acid extraction using ionic liquids. The ionic liquids can be used more than six times without losing its extraction efficiency.
a r t i c l e
i n f o
Article history: Received 22 March 2016 Received in revised form 6 May 2016 Accepted 7 May 2016
Keywords: Ionic liquids Naphthenic acid Thiocyanate anion Regeneration Extraction
a b s t r a c t Naphthenic acid (NA) is a toxic compound that exists in the effluent discharged from highly acidic oil refineries. The amount of NA present in acidic crude oil can be as high as 4 wt%. The complicated structure of NA poses a challenge for oil refineries in their effort to extract NA from the heavy crude oil in an economical and environmental friendly manner. In the current study the extraction of NA from highly acidic model oil by ionic liquid (ILs) was performed using 1,8-diazobicycloundec-7-ene (DBU) based cation in combination with the thiocyanate anion. A detailed computer simulation study on the mechanism of NA extraction by the ILs was also performed. The extracted NA was completely recovered and the ILs used were regenerated by simple addition of water. It was found that increasing of the alkyl chain length increases the percent NA removal. Computer simulation suggests those thiocyanate anions are found to be playing a major role in the NA extraction process. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction The increase in the processing of acidic crude by refineries had raised significant environmental and operational concerns [1,2]. The major problem associated with the acidic crude oil is the presence of NA, which is considered as both toxic and corrosive. NA is generally described by the general formula CnH2n+zO2, where n indicates the number of carbon atoms; z (negative integer or zero) indicates the deficiency of hydrogen because of the presence of cyclic or aromatic groups [3–5]. NA is highly toxic to aquatic organ⇑ Corresponding author at: Center for Biofuel and Biochemical Research, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610 Tronoh, Perak, Malaysia. E-mail address: [email protected] (L.K. Chellappan). http://dx.doi.org/10.1016/j.fuel.2016.05.041 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
isms and is mostly present in the effluent discharge from different petrochemical industries processing acidic crude oil. Most of the studies focus on the NA toxicity in the effluent discharge. In order to avoid the toxic effect of NA acid on the aquatic system, a simple and efficient process for the extraction of NA from its source of origin (acidic crude oil) would be beneficial [6–8]. NA have higher solubility in water compared to hydrocarbons thus making it a major threat to aquatic environment [9]. The presence of NA can be as high as 4 wt% in heavy crude oil. The NA can also cause corrosion in the oil refineries and it can adversely affect the storage and combustion properties of the final product. On the other hand NA is an important raw material (2500–3500 USD/ton) having potential application in different areas such as production of paint and cross linking of rubber [10]. Hence, the removal and recovery of NA are
580
S.N. Shah et al. / Fuel 181 (2016) 579–586
highly advantageous to both the environment and to the industry. The presence of NA in crude oil is measured using the total acid number (TAN). TAN values greater than 0.5 mg KOH/g of crude oil are considered as potentially corrosive for refineries units [11]. The current method used in the industry for the extraction and recovery of NA is the soda wash method, which leads to the spillage of NA in the water reservoirs [10]. Adsorption using ion exchange and liquid–liquid extraction (LLE) using organic solvents are the methods that have been tested for the recovery of NA from crude oil at lab scale [12–14]. Decarboxylation is another alternative for acidity reduction of highly acidic crudes [15,16]. Esterification using SnO/Al2O3 had been used for diesel streams as well as for crude oil [17,18]. However, all these methods have certain operational difficulties such as harsh reaction conditions, higher operational costs and lack of environmental viability. The restrictive factors in the current methods demands for a simple, environmental friendly and industrially viable process for the extraction of NA. Ionic liquids (ILs), solvents composed entirely of ions, have found application in energy, nuclear, catalysis, biomass, electrochemical and solar cells industries [19–26]. Although different ILs are used for extraction of NA, most of them are applicable only in the case of very low total acid number (<0.5 TAN). Moreover, most of them are using large amount of volatile organic solvents and the regeneration procedure of ILs and NA is quite complicated [27–31]. It has been shown that using anion with high pKa can increase the extraction efficiency of NA [28,30]. This essentially suggests that the more basic the IL is, the higher extraction efficiency will become. However the effect of the basicity of the cation has not been tested. It has also been proposed that the mechanism of NA extraction using ILs occur through the formation of a cage-like structure of the IL species around the NA molecules [30]. However, no details study about the mechanism has been performed to date. In this work, the extraction of NA using diazabicyclo undecane (DBU) based cation and thiocyanate anion is studied. The DBU based cation was selected for the extraction studies because conjugate acid of DBU is more basic than the conjugate acid of imidazole (>11 vs 7) [34,35]. In theory the DBU-based cation should have higher extraction efficiency. The thiocyanate anion is incorporated because of the presence of two coordination site, which will lead to the enhanced interaction between ILs and NA. In addition, computational study is performed to understand the molecular level mechanism of the extraction process. 2. Experimental 2.1. General information The chemicals used in this study were purchased from Sigma– Aldrich (Bornem, Belgium) and Acros Organics (Geel, Belgium). The synthesis and characterization of the ILs used in this study are performed using previously reported procedure [32]. The ILs used in this study are 1-hexyl-1,8-diazobicycloundec-7-ene thiocyanate, [DBU] [Hex], 1-octyl-1,8-diazobicycloundec-7-ene thiocyanate, [DBU] [Oct], 1-decyl-1,8-diazobicycloundec-7-ene thiocyanate, [DBU] [Dec]. The NMR spectra are recorded using Bruker Advance 500 spectrometer (operating at 500 MHz for 1H and 125 MHz for 13C). Number of scans is 16 and the pulse programme used is zg30. 2.2. Preparation of model acid oil In the current study highly acidic oil is used for de-acidification. The total acid number of the acidic oil used in this study is 3.46 (±0.01) mg KOH/g of oil, which is calculated using Metler Toledo
Auto titrator according to the ASTM D664 method. The acidic oil is prepared by adding commercial NA to dodecane. Dodecane is used in this study because of its closed proximity with kerosene, jet fuel oil and diesel. All these streams have the maximum concentration of naphthenic acid [33]. 2.3. Deacidification process The de-acidification of acidic oil is done by mixing model oil and ILs in a round bottom flask at constant stirring for one hour. The temperature is adjusted using silicon bath controlled by a hot plate and thermocouple. Once the reaction has completed, the reactants are transferred into a separating funnel. The two layers were easily separated because of the density difference. The de-acidified model oil is collected from the top of the separating funnel and its total acid number is analyzed. The percent NA is calculated using the formula given below:
Percent Naphthenic Acid Remov al ¼
1
TANf TANi
100;
where TANf and TANi refer to final and initial TAN of the oil respectively. 2.4. Computational details In order to understand the mechanism of the naphthenic acid extraction with DBU/SCN-based ILs, molecular dynamics simulation was performed on the [DBU-Hex] [SCN] system. To obtain reasonable statistics, the system is comprised of 1527 n-dodecane molecules, 86 [DBU-Hex] [SCN] ILs molecules, and 10 cyclohexyl acetic acid as the model NA. The simulation is performed using the AMBER [34] molecular dynamics package employing the GAFF-force field [35]. The charges were obtained using the RED Server Tools [36–39]. One modification was made to the angular force constant of the SACAN. The modification was made due to the unmodified GAFF force constant causing the SCN molecule to be mostly in the bent form, while from our geometry optimization with ab initio method suggests that the SCN molecule is in linear form at equilibrium, and thus should be in the linear form most of the time. A molecular dynamics simulation on the pure [DBUHex] [SCN] system gives a density value of 1.0419 g/mL, which is in good agreement with the experimental density of 1.0564 g/mL obtained in our lab. The percent error is calculated to be 1.3726%. The initial configuration of the system namely the topology as well as the input coordinate files required for the simulation are obtained using Avogadro [40], Packmol [41] and AmberTools 14 [42] software packages. The molecules are fitted into a cubic box using Packmol. The initial configuration of the system is minimized using the steepest decent method for 30,000 steps. Next, the system is equilibrated in the constant pressure-constant temperature ensemble (NPT) using Langevin dynamics with a time step of 0.5 fs, and collision frequency of 1.0 ps1. The temperature of the system is set to 300 K while the heat bath temperature coupling was set to 1.0 psi. The pressure of the system is set to 1 bar with isotropic position scaling while the pressure relaxation time was set to 1.0 psi. Periodic boundary condition is used to account for the boundary effect. The Van der Waals cutoff and the direct space-inverted space Particle-Mesh Ewald cutoff are set to 10 Å. The equilibration step was performed for 150 ns. The system is set to have reached equilibrium by making sure that the density and the potential energy of the system were stable. The plots of the density and the potential energy of the system are shown in Figs. 1 and 2 in the supporting information.
581
S.N. Shah et al. / Fuel 181 (2016) 579–586
3. Results and discussions 3.1. Effect of ionic liquid/oil ratio The most important parameter in the extraction study is the ILs to oil ratio. In this study three different ILs having different alkyl chain length are used. The ILs used in this study are shown in Fig. 1. The effect of different IL/oil ratio is investigated by varying the amount of ILs. The effect of IL/oil ratio is given in Table 1 and Fig. 2. The increase in the IL/model oil ratio increased the percent of NA extracted for all ILs used in the study. The extraction efficiency of the ILs follows the order [DBU-Dec] [SCN] > [DBU-Oct] [SCN] > [DBU-Hex] [SCN]. Longer the alkyl chain length of ILs, greater the extraction efficiency. When the extraction is performed at room temperature, [DBU-Dec] [SCN] is found to be the most effective in NA extraction. [DBU-Dec] [SCN] extracted about 35% NA at an IL/model oil ratio of 0.010, while [DBU-Oct] [SCN] extracted 32% NA. The lowest extraction performance is observed for [DBUHex] [SCN] (8.83%). The [DBU-Oct] [SCN] and [DBU-Dec] [SCN] showed similar extraction performance until IL/oil ratio of 0.03. When IL/oil ratio exceeded 0.03, the extraction performance between [DBU-Oct] [SCN] and [DBU-Dec] [SCN] increased significantly but showed different values. For instance, [DBU-Dec] [SCN] showed 100% NA extraction at an IL/model Oil of 0.06, while [DBU-Oct] [SCN] showed only 91% extraction performance at similar IL/model oil ratio. The reason for this difference could be due to the increased Van der Waals interaction between the NA and the longer alkyl chain in [DBU-Dec] [SCN] and the higher miscibility of n-dodecane with [DBU-Dec] [SCN] [31]. Among the ILs studied, [DBU-Hex] [SCN] showed the lowest extraction performance. For example, [DBU-Hex] [SCN] showed 100% extraction performance at an IL/model oil ratio of 0.10. As reported earlier [DBU-Oct] [SCN] and [DBU-Dec] [SCN] showed 100% extraction efficiency at a lower IL/model Oil ratio. The lower extraction performance of [DBU-Hex] [SCN] is related to its shorter chain length and higher viscosity [43]. The shorter alkyl chain length reduces the extent of Van der Waals interaction with the NA and the higher viscosity prevents the effective mass transport of the NA moiety. The imidazolate anion based ILs reported by Yu Sun et al. with [C8mim]+ cation showed 100% extraction efficiency at an IL/model
Table 1 Effect of the IL/oil ratio on naphthenic acid removal. IL/oil ratio (wt%)
0.010 0.020 0.030 0.040 0.050 0.06 0.07 0.08 0.09 0.1
100
Percent Naphthenic Acid Removal
After the system has attained equilibrium, the production run is performed in the NPT ensemble for 10 ns, using similar parameters as for the equilibration run. The trajectory of the 10 ns production is analyzed to study the mechanism of the naphthenic acid extraction by the ionic liquid.
Percent NA removal [DBU-Hex] [SCN]
[DBU-Oct] [SCN]
[DBU-Dec] [SCN]
8.83 21.32 35.09 46.98 57.17 66.12 75.24 82.99 91.76 100
32.39 48.74 63.76 76.26 85.61 91.31 99.85 100 100 100
34.93 52.51 67.86 82.26 93.80 100 100 100 100 100
[DBU-Hex][SCN] [DBU-Oct][SCN] [DBU-Dec][SCN]
80
60
40
20
0 0.02
0.04
0.06
0.08
0.10
IL/Oil ratio (W/W) Fig. 2. Effect of IL/oil ratio on percent naphthenic acid removal (j, [DBU-Hex] [SCN]; d, [DBU-Oct] [SCN]; N, [DBU-Dec] [SCN]) (stirring rate = 500 RPM, reaction time = 1 h).
oil ratio of 0.008 with a total acid number of 0.20 [30]. This extremely low total acid number do not even satisfied the criteria of acidic crude oil, where the total acid number should be greater than 0.5 mg KOH/g of oil [33]. In one of other studies [C8mim]+ cation with phenolate anion showed 100% extraction efficiency at an IL/oil ratio of 0.09 and total acid number of the model oil is 1.44 mg KOH/g of oil [27]. In comparison to the work reported by Yu Sun et al. [DBU-Oct] [SCN] requires higher IL/model ratio (0.07) for complete de-acidification of the model oil. One of the reasons for this relatively high IL/oil ratio is the high acidic oil used in the current study that having a total acid number of 3.46 mg KOH/g of oil. The thiocyanate based ILs perform exceptionally well as compared to N-alkyl pyridinium bromides and 1-alky-3methylimidazolium bromides [31]. For instance, 1-octyl-3methylimidazolium bromide, [C8mim] [Br], extracted only 93% of NA from a model oil with an IL/model oil ratio of 0.1, even if the acidity of the model system was very low (<0.5 TAN). At the similar conditions, 1-octylpyridinium bromide extracted less than 65% of NA. 3.2. Effect of temperature on naphthenic acid extraction
Fig. 1. Overview of ionic liquids used in this study.
The effect of temperature on the de-acidification is investigated using [DBU-Dec] [SCN] in a temperature range of 303.15–383.15 K. [DBU-Dec] [SCN] is used because of its higher extraction efficiency. Other extraction parameters are kept constant and the results are shown in Fig. 3. It is found that temperature do not have a significant influence on the extraction efficiency of ILs.
582
S.N. Shah et al. / Fuel 181 (2016) 579–586 100
Percent Naphthenic Acid Removal
Percent Naphthenic Acid Removal
72 71 70 69 68 67 66 65
90
80
70
60
50 300
320
340
360
1
380
2
3
4
Temperature
Only three percent enhancement was observed when the temperature was increased from 303.15 K to 383.15 K. The extraction performance slightly increased when the temperature was increased from 303.15 K to 360.15 K. However, there was no increase in extraction efficiency when the temperature increased above 360.15 K. The small increment in the extraction performance with the increase in temperature can be attributed to the enhanced mass transport due to the lowering of the viscosity of the ILs at high temperature. The 1H NMR spectrum of the ILs was recorded after the extraction process in order to confirm the stability of the ILs under elevated temperature. The 1H NMR spectra showed no sign of decomposition of the ILs used (Fig. 3 supporting information). 3.3. Recyclability of ionic liquids The recyclability of the materials used in a process is the most important parameter determining its sustainability and industrial viability. The extent of reusability of the ILs would be a reflection of the sustainability of the current process. The reusability of ILs is investigated using [DBU-Dec] [SCN]. An IL/oil ratio of 0.05 is used and the extraction is performed at room temperature. A model oil system having a TAN strength of 0.6 mg KOH/g of oil is used. The model oil is completely separated from the ILs after each extraction and the percentage of NA removal is calculated by measuring the TAN. The procedure is repeated by using fresh model oil and the previously used ILs. The experiment is repeated until the NA extraction efficiency of the ILs is reduced to less than 60% and the results are shown in Fig. 4. It is clear from the experimental results that the DBU based ILs can be recycled more than six times without significant reduction in their extraction efficiency. The extraction efficiency remained more than 95% even after seven cycles of extraction and DBU derived ILs showed more than 50% extraction efficiency even after they have been used more than 10 times. 3.4. Effect of different total acid number of oil The different TAN strengths of the model oil can also affect the extraction efficiency. The efficiency of the DBU derived ILs for the deacidification of model oil with different TAN strength is investigated by preparing model oil system with TAN up to 8 mg KOH/g. The study is performed using [DBU-Dec] [SCN] with the optimized IL/oil ratio of 0.05 and the results are shown in Fig. 5. The TAN
6
7
8
9
10
Fig. 4. Recyclability study of ionic liquids (j, [DBU-Dec] [SCN] with 0.05 IL/oil ratio at 303 K) (stirring rate = 500 RPM, reaction time = 1 h).
100
Percent Naphthenic Acid Removal
Fig. 3. Effect of temperature on naphthenic acid removal (j, [DBU-Dec] [SCN] with 0.03 IL/oil ratio) (stirring rate = 500 RPM, reaction time = 1 h).
5
No of Reactions
90 80 70 60 50 40 30 20
1
2
3
4
5
6
7
8
Total Acid Number Fig. 5. Effect of increase in total acid number on extraction efficiency of naphthenic acid (j, [DBU-Dec] [SCN] with IL/oil ratio of 0.05; (stirring rate = 500 RPM, reaction time = 1 h).
strength of the model oil is increased by adding more NA to the model oil. The extraction and separation time is set at one hour. The [DBUDec] [SCN] is able to completely de-acidify the model oil with a TAN value of 3.0 mg KOH/g. Although the TAN value of the model oil is increased significantly (8 mg KOH/g), the DBU derived ILs showed a deacidification efficiency of more than 40%, which indicates their potential in deacidification of highly acidic model systems.
3.5. Recovery of naphthenic acid Besides de-acidifying crude oil, the recovery of NA after the extraction with ILs is important due to its potential utilization in the chemical industry. Hence, a byproduct could be produced and it will lead to better economic potential for the proposed extraction process. The recovery of the NA from the ILs was studied for the purpose of exploring the extent of possible NA to be recovered as a value added product. The NA is recovered by adding deionized water into the ILs layer. Two layers were formed because of the hydrophobic nature of the NA and the hydrophilic nature of the ILs used ([DBU-Hex]
583
S.N. Shah et al. / Fuel 181 (2016) 579–586
[SCN]). The NA is separated from the IL layer in water. After this, NA is dried under vacuum and subjected to FTIR (Fig. 4 in supporting information) and 1H NMR analysis (Fig. 5 in supporting information). Both 1H NMR and FTIR spectrum was found to be similar to that of the pure NA. The FTIR spectrum of the pure NA showed the presence of several important functional groups that characterize it. The peak at 1698 cm1 confirmed the presence of carbonyl group (AC@O) whilst the strong peak at 1258 cm1 represents the CAO stretch in carboxylic acid. The broad peak at 2956 cm1 corresponds to the CAH stretch and another peak at 1463 cm1 is due to the CAH bend in alkanes. More than 97% of the NA appeared in the model oil was recovered by the addition of water. The ILs used for NA extraction was regenerated by the evaporation of the water using a rotary evaporator and the 1H NMR spectra of the recovered ILs was similar as in the case of pure ILs (Fig. 6 in supporting information). 3.6. Molecular dynamics simulation of ionic liquid, naphthenic acid, and model oil There were 3 possible mechanisms by which the NA could be removed by ILs namely: (i) the extraction of the acid by the anionic species; (ii) the formation of liquid clathrate in the sustaining medium with cage-like structure of the ILs surrounding the NA; (iii) the p–p interaction between the NA and the cationic species [30]. The molecular dynamics simulation performed in this study is based on classical mechanics nature and thus will not reveal any p–p interaction phenomenon between the NA and the DBU-based cation, which is of electronic nature. Nevertheless, the molecular dynamics simulation should be able to model the extraction of NA by the anionic species, and the formation of cage-like structure of the ionic liquids around the naphthenic acid. To look for the cage-like structure, the relative position of the DBU-cation, thiocyanate anion, and the model oil around the NA molecule as their trajectory is integrated over time is determined.
To simplify the analysis, N1-nitrogen atom of the DBU-cation, the C-carbon atom of the thiocyanate anion, and the C, C5, C6, and C11-carbon atom of the n-dodecane model oil are used as the reference as shown in Fig. 6. Fig. 7 shows the radial distribution function (RDF) plot of all the ILs and model oil molecules around a representative NA. Based on the plot, it can be deduced that the first shell surrounding the NA can be complicated with three different species surrounding the NA, as evidenced by the two extra shoulders on the first peak. Fig. 8 shows the radial distribution function (RDF) plot of the DBU-cation, the SCN anion, and the n-dodecane model oil from the geometrical center of the cyclohexyl acetic acid molecule. The radial distribution function (RDF) plot of all the 10 types of NA showed that the average distance from the center of geometry for the NA for the N1-nitrogen on DBU-cation is about 7.1 Å, While the average distance to the nearest thiocyanate anion (carbon atom) is on average 5.4 Å and average distance to the nearest ndodecane model oil is 5.7 Å. The results suggest that the NA has higher tendency to be surrounded by the model oil molecules first, followed by thiocyanate anions and the least is DBU cations since it is the furthest from the center of geometry of the NA. This might suggest that the DBU-cation is just a spectator ion attracted to the thiocyanate anion and not adjacent to the NA, thus forming a cage-like structure as suggested. In addition, the model oil is also very close to the NA molecule, suggesting that if there is any cage-like structure formed within the system, it also includes the model oil in it. A better conclusion is that there is no cage-like structure formed by the cation and the anion around the NA as suggested previously. From the integration of the first RDF peaks, the number of DBUcation surrounding the naphthenic acid molecule varies between 1 and 3, whilst for the thiocyanate anion it is between 1 and 2, and for the n-dodecane model oil surrounding the NA it is between 2 and 5. The reasonably high number of DBU-cation suggests there is a possibility that the cation could form a cage-like structure
C5 C
C11 C6
OH S
C5
N
C
O
carbonyl
N
cyclohexyl N1
N
Fig. 6. The position of the atoms used in the simulation.
584
S.N. Shah et al. / Fuel 181 (2016) 579–586
36
1.2
32
1.0
DBU-cation SCN-anion Dodecane
28
0.8
24 20
g (r)
g (r)
0.6
0.4
16 12 8
0.2
4
0.0 0
0
3
6
9
12
15
18
0
21
5
10
15
20
Distance (Angstrom)
Distance (Angstrom) Fig. 7. The RDF of ionic liquid and the model oil around a representative naphthenic acid.
Fig. 9. The RDF of the DBU-cation, SCN-anion, and the n-dodecane model oil from the center of geometry of the carbonyl moiety of the naphthenic acid.
12
16
DBU-cation SCN-anion Dodecane
14 12
DBU-cation SCN-anion Dodecane
10
8
g (r)
g (r)
10 8 6
6
4
4 2
2 0
0 0
4
8
12
16
20
Distance (Angstrom)
0
4
8
12
16
20
Distance (Angstrom)
Fig. 8. The RDF of each of the DBU-cation, SCN-anion, and n-dodecane around a representative naphthenic acid.
Fig. 10. The RDF of the DBU-cation, SCN-anion, and the n-dodecane model oil from the center of geometry of the C5-cyclohexyl moiety of the naphthenic acid.
around the NA. However, the low number of thiocyanate anion, and the high number of n-dodecane model oil supports the notion that the cage like structure does not exist. Although the DBU-cation molecules are close enough to form cages surrounding the NA molecules, the distribution of the DBUcation is mostly near the polar side of the NA. This is evidenced from Figs. 9 and 10. Fig. 7 depicts the RDF plot of the center of geometry for the carbonyl moiety (CAOAO) to the N1-nitrogen of the DBU-cation, to the C-carbon of the thiocyanate anion, and to the terminal and inner carbon (C, C5, C6, C11) of the n-dodecane model oil. The average distance from the center of geometry of the carbonyl moiety to the DBU-cation is 5.1 Å, while the same average distances to the C-carbon of the thiocyanate anion and to the n-dodecane model oil are 4.8 and 4.9 Å respectively. Fig. 10 shows the RDF of the C5-carbon of the cyclohexyl ring moiety of the NA to the N1-nitrogen of the DBU-cation, to the C-carbon of the thiocyanate anion, and to the terminal and inner carbon (C, C5, C6, C11) of the n-dodecane molecule. The average distance from the C5-carbon to the N1-nitrogen of the DBU-cation is 8.8 Å, while the average distance to the C-carbon of the thiocyanate anion is 7.4 Å, and the average distance to the n-dodecane model oil is 4.7 Å. From these two figures, it is evidenced that only the carbonyl part of the NA is likely to be surrounded by the DBU-cation and the
thiocyanate anion, while the cyclo hexyl part of the NA is only surrounded by the n-dodecane model oil. Fig. 11 shows a snapshot of one of the NA molecules from the simulation. From these figures, it is evidenced that while the DBU-cation and the thiocyanate anion do surround the NA, they are only located on the polar side of the NA (i.e. the carbonyl moiety), whereas the cyclohexyl ring is mostly surrounded by the n-dodecane model oil. These evidences do not support the idea of the cage-like structure formation around the NA. Based on the presented argument above, the most likely interaction between the ILs and the NA is through the hydrogen bonding between the OH–hydrogen of the NA to the anion of the ILs. To confirm the interaction mechanism involving the OH–hydrogen of the acid, the RDF plot of the OH–hydrogen to the sulfur and nitrogen atoms on the thiocynate anion is generated. Furthermore, to investigate probable interaction of the OH–hydrogen with the nitrogen atoms on the DBU-cation, the RDF plot of the nitrogen atoms on the DBU-cation around the OH–hydrogen is also generated. As evidenced from the RDF plot in Fig. 12, the average distance from the OH–hydrogen to the nitrogen atom and the sulfur atom on the thiocyanate anion is 1.77 and 1.97 Å respectively, while the distance from the OH–hydrogen to the N1-nitrogen and N-nitrogen on the DBU-cation is 4.66 and 4.76 Å respectively.
S.N. Shah et al. / Fuel 181 (2016) 579–586
585
thiocyanate anion at any one time. The coordination to one thiocyanate anion at both the sulfur and the nitrogen at the same time is not observed, although this might be because of the low number of NA to thiocyanate anion ratio used in the simulation that makes the probability of the OH–hydrogen to attach to another thiocyanate anion is higher compared to sharing the thiocyanate anion by two OH–hydrogen atoms. 3.7. Mechanism of naphthenic acid extraction
Fig. 11. A snapshot of the naphthenic acid (pink) is being surrounded by the DBUcation (green), SCN-anion (white), and the n-dodecane model oil (gray). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
80 70
SCN-Sulfur SCN-Nitrogen DBU-N-Nitrogen DBU-N1-Nitrogen
60
g (r)
50 40 30 20 10 0 0
4
8
12
16
20
Distance (Angstrom) Fig. 12. The RDF of the thiocyanate sulfur, thiocyanate nitrogen, DBU-N-nitrogen, and DBU-N1-nitrogen around the OH–hydrogen.
From the experimental results, the ILs with DBU-cation and thiocyanate anion displayed better performance than other previously published ILs used for extraction of NA. It was claimed that the pKa of the anion plays an important role in NA extraction [27,30]. However, from our results, it is evidenced that the pKa is not of significance. A comparison between the use of phenolate anion which has a pKa value of 10 [27], imidazole-based anion used in the previous study [30] had a pKa value higher than 10. However, the pKa of thiocyanate anion is around 1 [44]. If the pKa of the anion plays a big role in ILs extraction of NA, the DBU-cationthiocyanate anion ILs would have had a lower NA extraction values. However, as evidenced by our results, the NA extraction is higher with the DBU-cation-thiocyanate anion ILs. In addition to that, the pKa describes the ability of a negatively charge species to abstract protons from a hydrogen donor. In the NA extraction process by ILs, the anion does not need to abstract the proton from the acid. Instead, the anion only needs to coordinate to the NA thus pulling the acid out from the model oil. From our results, we suspect that the number of available interaction site is the main factor causing the high NA extraction. Not only the thiocyanate anion has lower molecular weight thus increasing the number of thiocyanate anion for each gram of the ILs used in the extraction, it also has two coordination sites for each thiocyanate molecule. This doubles the coordination probability by the ILs and thus increasing the number of NA molecule extracted from the model oil during each extraction cycle. It has also been suggested that the cation and the alkyl side chain plays a role in extracting the NA [27,30,31]. On the contrary our simulation results show that the cation does not get close to the NA molecule thus rendering it only as a spectator in the whole process. It can be postulated that the alkyl chain on the cation increases the ability of the polar cation to be dissolved in the non-polar model oil thus increasing the probability of the anion to interact with the NA molecule. More studies are needed to provide the exact reason of the increase in the NA extraction with the increase in the alkyl chain length. 4. Conclusion
Given the much shorter average distance between the OH–hydrogen to the sulfur and nitrogen atoms on the thiocynate anion, it suggests that the NAAOH interacts only with the thiocyanate anion and not the DBU-cation. Hence it can be deduced for all the 10 NA molecules, the OH– hydrogen interacts dominantly with the nitrogen or the sulfur atoms of the thiocyanate anion. The site of the coordination preferred by the OH–hydrogen is the sulfur atom on the thiocyanate, as evidence by the integration of the first RDF peak, with the sulfur being on average 70% of the time and the nitrogen being on average 30% of the time coordinated to the OH–hydrogen. The thiocyanate has two coordination sites to form hydrogen bonding with the OH–hydrogen from the NA. The two coordination sites on the thiocyanate anion is contrary to the other anions reported and this might have provided the advantage for the thiocyanate anion over others used in the ILs extraction of NA. Nevertheless only one NA is attached to one
The thiocyanate based ILs are very effective in the deacidification of high TAN model oil with an extremely low IL/oil ratio. DBU based cation with alkyl groups such as hexyl, octyl, and decyl were used to study the effect of alkyl groups on extraction efficiency. The extraction efficiency of the DBU-based thiocyanate ILs increases with the increase in the alkyl spacer length on the DBU cation. [DBU-Dec] [SCN] showed maximum extraction efficiency. However, simulation study using the DBU-based cation suggests that it is not directly coordinated to the NA in the extraction process. Instead, the thiocyanate anion is the species coordinated directly with the NA. The simulation study showed that thiocyanate anion has two possible coordination sites. This increases the capability of the thiocyanate anion to interact with the NA from the oil, thus increasing the ability of the thiocyanate-based ILs to abstract NA. All the ILs can be reused more than six times for the extraction without compromising their
586
S.N. Shah et al. / Fuel 181 (2016) 579–586
extraction performance. More than 98% of the extracted NA was regenerated by the simple addition of water into the ILs. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fuel.2016.05.041. References [1] http://www.apecconsulting.com/PDF/GlobalHighTAN.pdf. [2] Pedro P, Roger M, Luis Z. AquaconversionTM: a new option for residue conversion and heavy oil upgrading. J Vis Technol 1998;6:5–13. [3] Fan TP. Characterization of naphthenic acids in petroleum by fast atom bombardment mass spectrometry. Energy Fuel 1991;5:371–5. [4] Hsu CS, Dechert G, Robbins W, Fukuda E. Naphthenic acids in crude oils characterized by mass spectrometry. Energy Fuel 2000;14:217–23. [5] Rudzinski WE, Oehlers L, Zhang Y, Najera B. Tandem mass spectrometric characterization of commercial naphthenic acids and a Maya crude oil. Energy Fuel 2002;16:1178–85. [6] Thomas K, Langford K, Petersen K, Smith A, Tollefsen K. Effect-directed identification of naphthenic acids as important in vitro xeno-estrogens and anti-androgens in North Sea offshore produced water discharges. Environ Sci Technol 2009;43:8066–71. [7] Giesy JP, Anderson JC, Wiseman SB. Alberta oil sands development. Proc Nat Acad Sci 2010;107:951–2. [8] Frank RA, Fischer K, Kavanagh R, Burnison BK, Arsenault G, Headley JV. Effect of carboxylic acid content on the acute toxicity of oil sands naphthenic acids. Environ Sci Technol 2008;43:266–71. [9] Headley JV, McMartin DW. A review of the occurrence and fate of naphthenic acids in aquatic environments. J Environ Sci Health Part A 2004;39:1989–2010. [10] Brient JA, Wessner PJ, Doyle MN. Naphthenic acids. Kirk-Othmer Encyclopedia of Chemical Technology; 1995. [11] Testing ASf, Materials. ASTM D-664: standard test method for acid number of petroleum products by potentiometric titration. Annual book of ASTM standards; 1989. [12] Holowenko FM, MacKinnon MD, Fedorak PM. Characterization of naphthenic acids in oil sands wastewaters by gas chromatography–mass spectrometry. Water Res 2002;36:2843–55. [13] Gaikar V, Maiti D. Adsorptive recovery of naphthenic acids using ion-exchange resins. React Funct Polym 1996;31:155–64. [14] Hoeiland S, Barth T, Blokhus A, Skauge A. The effect of crude oil acid fractions on wettability as studied by interfacial tension and contact angles. J Pet Sci Eng 2001;30:91–103. [15] Ding L, Rahimi P, Hawkins R, Bhatt S, Shi Y. Naphthenic acid removal from heavy oils on alkaline earth-metal oxides and ZnO catalysts. Appl Catal A 2009;371:121–30. [16] Zhang A, Ma Q, Wang K, Liu X, Shuler P, Tang Y. Naphthenic acid removal from crude oil through catalytic decarboxylation on magnesium oxide. Appl Catal A 2006;303:103–9. [17] Wang YZ, Sun XY, Liu YP, Liu CG. Removal of naphthenic acids from a diesel fuel by esterification. Energy Fuel 2007;21:941–3. [18] Wang Y-z, Li J-y, Sun X-y, Duan H-l, Song C-m, Zhang M. Removal of naphthenic acids from crude oils by fixed-bed catalytic esterification. Fuel 2014;116:723–8. [19] Shah SN, Lethesh KC, Mutalib M Abdul, Pilus M, Binti R. Extraction and recovery of naphthenic acid from acidic oil using supported ionic liquid phases (SILPs). Chem Prod Proc Model 2015;10:221–8.
[20] MacFarlane DR, Tachikawa N, Forsyth M, Pringle JM, Howlett PC, Elliott GD. Energy applications of ionic liquids. Energy Environ Sci 2014;7:232–50. [21] Ullah Z, Bustam MA, Man Z. Biodiesel production from waste cooking oil by acidic ionic liquid as a catalyst. Renew Energy 2015;77:521–6. [22] Ohno H. Electrochemical aspects of ionic liquids. Wiley.com; 2011. [23] Rao C, Venkatesan K, Nagarajan K, Srinivasan T. Dissolution of uranium oxides and electrochemical behavior of U (VI) in task specific ionic liquid. Radiochim Acta 2008;96:403–9. [24] Swatloski RP, Spear SK, Holbrey JD, Rogers RD. Dissolution of cellose with ionic liquids. J Am Chem Soc 2002;124:4974–5. [25] Wu B, Reddy R, Rogers R. Novel ionic liquid thermal storage for solar thermal electric power systems. Sol Eng 2001:445–52. [26] Ping B, Yusof M. Characteristics and properties of fatty acid distillates from palm oil. Oil Palm Bull 2009;59:5–11. [27] Nasir Shah S, Chellappan L Kallidanthiyil, Gonfa G, Mutalib MIA, Pilus RBM, Bustam MA. Extraction of naphthenic acid from highly acidic oil using phenolate based ionic liquids. Chem Eng J 2016;284:487–93. [28] Nasir Shah S, Mutalib MIA, Pilus RBM, Lethesh KC. Extraction of naphthenic acid from highly acidic oil using hydroxide-based ionic liquids. Energy Fuels 2014;29:106–11. [29] Shi LJ, Shen BX, Wang GQ. Removal of naphthenic acids from Beijiang crude oil by forming ionic liquids. Energy Fuels 2008;22:4177–81. [30] Sun Y, Shi L. Basic ionic liquids with imidazole anion: new reagents to remove naphthenic acids from crude oil with high total acid number. Fuel 2012;99:83–7. [31] Duan J, Sun Y, Shi L. Three different types of heterocycle of nitrogen-containing alkaline ionic liquids treatment of acid oil to remove naphthenic acids. Catal Today 2013;212:180–5. [32] Lethesh KC, Shah SN, Mutalib MA. Synthesis, characterization, and thermophysical properties of 1,8-diazobicyclo [5.4. 0] undec-7-ene based thiocyanate ionic liquids. J Chem Eng Data 2014;59:1788–95. [33] Speight JG. Chapter 1 – Naphthenic acids in petroleum. In: Speight JG, editor. High acid crudes. Boston: Gulf Professional Publishing; 2014. p. 1–29. [34] Case D, Darden T, Cheatham III TE, Simmerling C, Wang J, Duke R. AMBER 12. San Francisco: University of California; 2012. p. 1. [35] Wang J, Wolf RM, Caldwell JW, Kollman PA, Case DA. Development and testing of a general amber force field. J Comput Chem 2004;25:1157–74. [36] Vanquelef E, Simon S, Marquant G, Garcia E, Klimerak G, Delepine JC. RED server: a web service for deriving RESP and ESP charges and building force field libraries for new molecules and molecular fragments. Nucl Acids Res 2011;39:W511–7. [37] Dupradeau F-Y, Pigache A, Zaffran T, Savineau C, Lelong R, Grivel N, et al. The RED. Tools: advances in RESP and ESP charge derivation and force field library building. Phys Chem Chem Phys 2010;12:7821–39. [38] Bayly CI, Cieplak P, Cornell W, Kollman PA. A well-behaved electrostatic potential based method using charge restraints for deriving atomic charges: the RESP model. J Phys Chem 1993;97:10269–80. [39] Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon MS, Jensen JH, et al. General atomic and molecular electronic structure system. J Comput Chem 1993;14:1347–63. [40] Hanwell MD, Curtis DE, Lonie DC, Vandermeersch T, Zurek E, Hutchison GR. Avogadro: an advanced semantic chemical editor, visualization, and analysis platform. J Cheminformatics 2012;4:17. [41] Martínez L, Andrade R, Birgin EG, Martínez JM. PACKMOL: a package for building initial configurations for molecular dynamics simulations. J Comput Chem 2009;30:2157–64. [42] Case D, Babin V, Berryman J, Betz R, Cai Q, Cerutti D. Amber 14; 2014. [43] Lethesh KC, Shah Syed Nasir, Mutalib MI Abdul. Synthesis, characterization, and thermophysical properties of 1,8-diazobicyclo[5.4.0]undec-7-ene based thiocyanate ionic liquids. J Chem Eng Data 2014;59:1788–95. [44] Martell A, Smith R. NIST stability constants of metal complexes database 46, 6.0.. Gaithersburg (MD): US Department of Commerce; 2001.