Purification of Indole Contained in Wash Oil by Combination of Extraction and Crystallization (Part 1: Recovery and Concentration of Indole Contained in Wash Oil by Solvent Extraction)

For the purpose of determining the high-purity purification of indole (IN) contained in wash oil in concentrations of 5.75%, this study, first of all, investigated the concentration of IN contained in wash oil using a combination of methanol extraction to recover IN in the wash oil, and n-hexane re-extraction to concentrate IN present in the extract phase, recovered through methanol extraction. In order to examine the effect of each operation on the recovery and concentration of IN contained in the wash oil, batch 1-stage and batch co-current 5-stage distribution equilibrium was performed simultaneously. As 5-stage equilibrium extraction and re-extraction progressed, the recovery rate of IN decreased to about 79.1%, while IN composition in the raffinate phase recovered by re-extraction was highly concentrated to about 73.3%. From the high recovery rate and composition of IN obtained through this study, we confirmed that the combination examined by this study was one of the very useful combinations for the concentration of IN present in the wash oil. Furthermore, we reviewed the recovery and concentration process of IN contained in the wash oil using the experiment results of the extraction and the re-extraction obtained from this study.


Introduction
Wash oil, which is a mixture oil of approximately 10-30 compounds, contains many valuable nitrogen-containing compounds (NCs), such as quinoline (QU), iso-quinoline, IN, and quinaldine. Among them, IN has been recognized as an intermediate for the productions of essential amino acids, such as tryptophan, alkaloid-based medicines, and perfumes, such as jasmine and neroli oil [1][2][3][4][5][6]. On the other hand, the NCs mentioned above are undesirable impurities in the wash oil because they contribute to air pollution and have unpleasant odors. At the present time, QU and the mixtures containing more than 0.1% QU are classified as hazardous chemicals (carcinogens, skin irritants, etc.). Therefore, wash oil containing about 1.38% QU is a hazardous chemical, so there are many restrictions on transportation and handling, etc. For the above reasons, the improvement of the quality of wash oil according to the reduction in NC and the review of the purification of useful components, such as IN using the crude NC mixture recovered through the quality improvement process, are considered to be very meaningful from the viewpoint of recycling of resources [1].
The composition of IN contained in wash oil varies depending on the manufacture company (about 3-7%), but the wash oil used in this study contained about 5.75% IN. As shown in Table 1, it can be seen that the boiling points of the compounds included in the washing oil are very close, so it is very difficult to separate each compound by distillation. It is believed that a multi-step procedure should be required to separate and purify of IN from the wash oil. The separation and purification of IN contained in the wash oil is thought  Up to the present, the crude separation of IN from the coal tar fraction recovered through distillation of the coal tar was investigated via separation operations, such as extraction using ionic liquids [7][8][9], supercritical extraction [10,11], liquid membrane permeation [4], azeotropic distillation, and traditional extraction methods [1][2][3]5,6,12] using various organic solvents, such as methanol, ethanol, formamide, etc. The purification of IN from IN-enriched distillate recovered through distillation of the extract was also investigated by an operation, such as inclusion complexation [13,14], adsorption [15], and crystallization [3,16].
In a previous study [3], we investigated the concentration of IN by the combination of extraction-distillation using a model coal tar fraction. A distillate containing about 54.3% IN was obtained by batch distillation of the extract phase, which was recovered by batch equilibrium extraction using an aqueous formamide solution as a solvent. Considering the results of a previous study, the low IN composition of the distillate and the high boiling point of formamide (483 K) used as an extraction solvent were raised as one problem.
In this study, the high concentration of IN contained in wash oil was experimentally investigated by a combination of extraction using an aqueous methanol solution, which makes it relatively easy to recover the solvent, as an extraction solvent, and re-extraction using n-hexane as a re-extraction solvent.

Experimental Method
A schematic diagram for extraction and re-extraction methods of a batch co-current 5-stage is presented in Figure 1 [1]. study, the above-mentioned eight kinds of commercial reagents were used w ther purification.

Experimental Method
A schematic diagram for extraction and re-extraction methods of a batch 5-stage is presented in Figure 1 [1].

Batch Equilibrium Extraction
To make a solvent of a certain concentration, methanol was mixed with t 1 L Erlenmeyer flask containing a certain amount of the feed (R0-wash oil, R nate phases recovered from extraction run of each stage) and the fresh solve aqueous methanol solution) were placed in a shaking water bath maintained a imental temperature and vibrated for a certain time to reach a liquid-liquid e After reaching the equilibrium, the mixture was settled for an aliquot of tim raffinate phase and the extract phase were separated using a 1 L separatory then the mass of each phase was measured. The separated raffinate phase solvent were used as the feed and a solvent of the next stage, respectively. T phases and the extract phases recovered from the equilibrium extraction run o were analyzed by adding a small amount of acetone, and their compositions mined. The analysis of two phases was carried out using a gas chromato Hewlett Packard Co., Houston, TX, USA, HP 6890), with a capillary column, H 0.32 mm I.D.) and equipped with a flame ionization detector (FID). The ana tions of the samples were as follows: carrier gas, N2; volume flow rate, 1 mL/m temperature, 523 K; sample volume, 1 µL; splitting ratio, 0.025; column te maintained at 383 K for 3 min, then increased at a rate of 5 K/min to 523 K, the to 593 K; detector temperature was set to 593 K [1].

Batch Equilibrium Re-Extraction
A 1 L Erlenmeyer flask containing a certain amount of the feed (R0-a m phase, which is a mixture of each extract phase recovered through methanol e extraction of batch co-current 5-stage, R1-R4-raffinate phases recovered thro extraction run of each stage) and the fresh solvent (E0-E4-n-hexane) were shaking water bath maintained at the experimental temperature and vibrated time to reach a liquid-liquid equilibrium. After that, the experimental meth used, and the GC analysis conditions of the samples are the same as those o extraction described above. Schematic diagram for equilibrium extraction and re-extraction methods of a batch cocurrent 5-stage [1]. Abbreviations are as follows: R, raffinate phase; E, extract phase; WO, wash oil; MEP, mixed extract phase; i, component i; 1, 2, 3, 4, 5, number of equilibrium extraction or re-extraction.

Batch Equilibrium Extraction
To make a solvent of a certain concentration, methanol was mixed with tap water. A 1 L Erlenmeyer flask containing a certain amount of the feed (R 0 -wash oil, R 1 −R 4 -raffinate phases recovered from extraction run of each stage) and the fresh solvent (E 0 -E 4 -aqueous methanol solution) were placed in a shaking water bath maintained at the experimental temperature and vibrated for a certain time to reach a liquid-liquid equilibrium. After reaching the equilibrium, the mixture was settled for an aliquot of time, and the raffinate phase and the extract phase were separated using a 1 L separatory funnel, and then the mass of each phase was measured. The separated raffinate phase and a fresh solvent were used as the feed and a solvent of the next stage, respectively. The raffinate phases and the extract phases recovered from the equilibrium extraction run of each stage were analyzed by adding a small amount of acetone, and their compositions were determined. The analysis of two phases was carried out using a gas chromatograph (GC, Hewlett Packard Co., Houston, TX, USA, HP 6890), with a capillary column, HP-1 (60 mL, 0.32 mm I.D.) and equipped with a flame ionization detector (FID). The analysis conditions of the samples were as follows: carrier gas, N 2 ; volume flow rate, 1 mL/min; injection temperature, 523 K; sample volume, 1 µL; splitting ratio, 0.025; column temperature, maintained at 383 K for 3 min, then increased at a rate of 5 K/min to 523 K, then 14 K/min to 593 K; detector temperature was set to 593 K [1].

Batch Equilibrium Re-Extraction
A 1 L Erlenmeyer flask containing a certain amount of the feed (R 0 -a mixed extract phase, which is a mixture of each extract phase recovered through methanol equilibrium extraction of batch co-current 5-stage, R 1 -R 4 -raffinate phases recovered through the reextraction run of each stage) and the fresh solvent (E 0 -E 4 -n-hexane) were placed in a shaking water bath maintained at the experimental temperature and vibrated for a certain time to reach a liquid-liquid equilibrium. After that, the experimental method, the GC used, and the GC analysis conditions of the samples are the same as those of methanol extraction described above.

Material System and Experimental Conditions
The material system and the experimental conditions according to each operation are summarized in Table 2. Volume fraction of water in solvent in initial state, y w,0 (-) 0.1-0.2 Volume ratio of fresh solvent to feed in initial state, E 0 /R 0 (-) 0.5-3 Volume ratio of fresh solvent to feed in initial state, E 0 /R 0 (-) 0.5-3 * raffinate phase recovered through extraction run of each stage (n = 1-5), ** mixed extract phase (∑ 5 n=1 E n ), which is a mixture of each extract phase recovered through extraction run of each stage (n = 1-5), *** raffinate phase recovered through re-extraction run of each stage (n = 1-5).

Batch Equilibrium Extraction
Wash oil and the raffinate phases recovered from each stage run were used as the feed, and an aqueous methanol solution was used as the fresh solvent. The operating temperature (T) and the volume of fresh solvent added to each stage (E 0 −E 4 ) are kept constant, and the liquid-liquid contact time (t), the number of equilibrium extraction (n), the volume fraction of water in the solvent in the initial state (y w,0 ), and the volume fraction of the fresh solvent to feed in the initial state (E 0 /R 0 ) are changed.

Batch Equilibrium Re-Extraction
The mixed extraction phase, which is a mixture of each extract phase recovered through a methanol extraction of n = 1-5, and each raffinate phase recovered in each stage of equilibrium re-extraction run using the above-mentioned mixed extraction phase were used as the feed of the re-extraction operation. Here, t, E 0 -E 4 , and T were kept constant, and n and E 0 /R 0 were changed.

Definition Equation
Here, m i,n , which is the distribution coefficient of the component i obtained by the extraction or re-extraction of the nth stage contact run, is defined as follows: where y i,n and x i,n denote the mass fraction of component i in the extract phase and the raffinate phase recovered from the extraction or re-extraction of the nth stage contact run, respectively.
Here, Y i,n , which is the yield of component i obtained from the extraction or reextraction of the nth stage, is defined as follows: where R 0 and x i,0 , respectively, represent the mass of the feed put into n = 1 contact run of the extraction or re-extraction, and the mass fraction of component i in the feed.
Additionally, E n and y i,n denote the mass of the extract phase and the mass fraction of Here, (β i,2MNA ) n or (β i,IN ) n , respectively, which are the selectivity of component i in reference to 2MNA or IN obtained from the extraction or re-extraction of the nth stage contact run respectively, are defined as follows:

Confirmation of Equilibrium Arrival Time
To confirm the time of reaching the equilibrium, the raffinate phases or the extract phases recovered through the extraction and re-extraction contact (a contact time of 12 h, 24 h, 72 h, and 96 h) between the feed and the fresh solvent were analyzed. The compositions of the raffinate phases and the extract phases recovered through contact for more than 48 h were almost the same, regardless of the contact time and the operation. Therefore, the liquid-liquid contact time was maintained at 72 h in the entire experiment of this study. Figure 2a presents the gas chromatogram of the wash oil used as the extraction feed of this study and the component names of the identified compounds. As mentioned above, they were identified through analysis by adding a small amount of six standard reagents purchased from Sigma-Aldrich Korea. As a result of GC identification, it can be seen that the wash oil contains relatively small amounts of two kinds of NCs (QU, and IN), whereas 2MNA, 1MNA, and BP, excluding NA are included in large amounts. The compositions of the 6 compounds quantified in this study are shown in Table 1. The composition of IN (peak No. 3), which is the aim component to concentrate in this study, was about 5.75%.

Recovery Performance of IN Contained in Wash Oil
In this study, in order to maximize the recovery of IN contained in the wash oil, methanol extraction of the wash oil was performed to investigate the effect of extraction factors and conditions on the recovery of IN. Figure 3 shows the effect of y w,0 on m i,1 and (β i,2MNA ) 1 obtained in n = 1 of the fixed extraction conditions (n = 1, E 0 /R 0 = 1, T = 303 K, t = 72 h). The m i,1 of NCs (i = QU, IN) with very high polarity were much larger than three kinds of bicyclic aromatic compounds (BACs; i = NA, 1MN, 2MN) and Bp, which were less than the polarity of NCs. Increasing y w,0 resulted in a decreasing m i,1 of each component because the polarity of the extract phase increased with an increase in its moisture in the extract phase. As can be seen from the definition of (β i,2MNA ) 1 and the result of m i,1 of NCs mentioned above, (β i,2MNA ) 1 of NCs, inversely, increased as y w,0 increased. This result suggested that the selectivity for the total NCs based on the total BACs contained in the wash oil could be very high. On the other hand, considering that the values of (β 1MNA,2MNA ) 1 , (i = 1MNA), and (β BP,2MNA ) 1 (i = BP) are almost 1, it was found that separation between these components (1MNA, 2MNA, and BP) by methanol extraction was difficult. In y w,0 = 0.1−0.  above, they were identified through analysis by adding a small amount of six standard 168 reagents purchased from Sigma-Aldrich Korea. As a result of GC identification, it can be 169 seen that the wash oil contains relatively small amounts of two kinds of NCs (QU, and IN), 170 whereas 2MNA, 1MNA, and BP, excluding NA are included in large amounts. The com-171 positions of the 6 compounds quantified in this study are shown in  Gas chromatograms of (a) extraction feed (wash oil), (b) raffinate phase (R 5 ) recovered through a methanol extraction run of n = 5, (c) re-extraction feed (mixed extract phase (∑ 5 n=1 E n ), which is a mixture of each extract phase recovered through a methanol extraction run of n = 1-5, (d) raffinate phase (R 5 ) recovered by an n-hexane re-extraction run of n = 5, and (e) mixed extract phase (∑ 5 n=1 E n ), which is a mixture of each extract phase recovered through an n-hexane extraction run of n = 1-5. Extraction experimental conditions are as follows: y w,0 = 0.1, E 0 /R 0 = 1, T = 303 K, n = 1-5, t = 72 h. Re-extraction experimental conditions are as follows: E 0 /R 0 = 0.5, T = 303 K, n = 1-5, t = 72 h. Peak number 1 was naphthalene (NA), 2 was quinolone (QU), 3 was indole (IN), 4 was 2-methylnaphthalene (2MNA), 5 was 1-methylnaphthalene (1MNA), and 6 was biphenyl (BP).

Recovery Performance of IN Contained in Wash Oil
In this study, in order to maximize the recovery of IN contained in the wash oil, methanol extraction of the wash oil was performed to investigate the effect of extraction factors and conditions on the recovery of IN.      Figure 5a,b refers to the w used as the feed in this study. As predicted by Figure 3, NCs with large polarity tracted into the extract phase, and it can be seen that the xi,n of NCs in the raffinate decreases rapidly as n increases. On the other hand, the other four components, asid the NCs, showed that xi,n values were almost constant regardless of n. This is beca compositions of the NCs contained in the wash oil are very low compared to three of BACs and BP, so even if a large amount of the NCs with large polarity is extra the equilibrium extraction proceed, the composition of BACs and BP according to t gress of n is not significantly affected. At n = 5, IN, which is the target componen cover in this study, was not contained in R5. Through this, the effect of recoverin the wash oil by methanol extraction could be confirmed. It can also be seen from 5b that the yi,n of the NCs decreases rapidly as the equilibrium extraction progress the yi,n of the other four kinds of components presented a tendency to be almost co In the ranges of n = 1-4, xi,n and yi,n of i = IN, namely xIN,n and yIN,n, showed the ran 0.2-3.3% and 1.63−16.0%, respectively. Considering that xIN,5 (i = IN) did not appea raffinate oil recovered at n = 5, it is expected that the quality of the wash oil will proved by recovering NC-free raffinate oil when using a multi-stage column unde mal extraction conditions using methanol. In all components of this study, m (βi,2MNA)n showed almost the same value regardless of n, so the change in the m (βi,2MNA)n according to n could not be recognized. Furthermore, (βIN,2MNA)n was abou gardless of n.   Figure 5a,b refers to the wash oil used as the feed in this study. As predicted by Figure 3, NCs with large polarity are extracted into the extract phase, and it can be seen that the x i,n of NCs in the raffinate phase decreases rapidly as n increases. On the other hand, the other four components, aside from the NCs, showed that x i,n values were almost constant regardless of n. This is because the compositions of the NCs contained in the wash oil are very low compared to three kinds of BACs and BP, so even if a large amount of the NCs with large polarity is extracted as the equilibrium extraction proceed, the composition of BACs and BP according to the progress of n is not significantly affected. At n = 5, IN, which is the target component to recover in this study, was not contained in R 5 . Through this, the effect of recovering IN in the wash oil by methanol extraction could be confirmed. It can also be seen from Figure 5b that the y i,n of the NCs decreases rapidly as the equilibrium extraction progresses, but the y i,n of the other four kinds of components presented a tendency to be almost constant. In the ranges of n = 1-4, x i,n and y i,n of i = IN, namely x IN,n and y IN,n , showed the ranges of 0.2-3.3% and 1.63−16.0%, respectively. Considering that x IN,5 (i = IN) did not appear in the raffinate oil recovered at n = 5, it is expected that the quality of the wash oil will be improved by recovering NC-free raffinate oil when using a multi-stage column under optimal extraction conditions using methanol. In all components of this study, m i,n and (β i,2MNA ) n showed almost the same value regardless of n, so the change in the m i,n and (β i,2MNA ) n according to n could not be recognized. Furthermore, (β IN,2MNA ) n was about 6, regardless of n. Figure 2b,c, respectively, show the gas chromatograms of the raffinate phase (R 5 ) recovered through a methanol extraction of n = 5, and a mixed extract phase (∑ 5 n=1 E n ) which mixed each extract phase recovered through a methanol extraction of n = 1−5. When the gas chromatogram of R 5 in Figure 2b was compared with that of the wash oil in Figure 2a, the peaks of NCs (peak numbers 2 and 3) did not appear in the R 5 , and the peak heights of the other compounds was almost the same as those of the wash oil. From the gas chromatogram of the mixed extract phase of Figure 2c, we could see that the peak height of each NC was slightly different from that of the wash oil. Two kinds of NCs were extracted with the progress of n, and the peak height of each NC slightly increased, but the peak heights of other compounds were almost the same.  mixed each extract phase recovered through a methanol extraction of n = 1−5. Wh gas chromatogram of R5 in Figure 2b was compared with that of the wash oil in Figu the peaks of NCs (peak numbers 2 and 3) did not appear in the R5, and the peak h of the other compounds was almost the same as those of the wash oil. From the gas matogram of the mixed extract phase of Figure 2c, we could see that the peak hei each NC was slightly different from that of the wash oil. Two kinds of NCs were ext with the progress of n, and the peak height of each NC slightly increased, but the heights of other compounds were almost the same.

Gas Chromatogram of Re-Extraction Feed (Mixed Extract Phase)
The mixed extract phase in Figure 2c mentioned above was used as the feed extraction in this study. The compositions of six kinds of compounds quantified study are shown in Table 1. The composition of the feed of IN, the target compone

Gas Chromatogram of Re-Extraction Feed (Mixed Extract Phase)
The mixed extract phase in Figure 2c mentioned above was used as the feed for reextraction in this study. The compositions of six kinds of compounds quantified in this study are shown in Table 1. The composition of the feed of IN, the target component for concentration through the re-extraction of this study, was about 1.08%.

Concentration Performance of IN Contained in Mixed Extract Phase
In order to concentrate the IN contained in the mixed extraction phase recovered by methanol extraction, this study performed a re-extraction operation using n-hexane as a solvent to examine the effect of re-extraction operation factors and conditions on the concentration of IN. Figure 6 presents the effect of E 0 /R 0 on m i,1 and (β i,IN ) 1 (based on IN) obtained in a n = 1 re-extraction run under the fixed re-extraction conditions (n = 1, T = 303 K, t = 72 h). In contrast to the above-mentioned methanol extraction, we found that the m i,1 and (β i,IN ) 1 of three kinds of BACs (NA, 2MNA, and 1MNA) and BP show significantly larger values than those of two kinds of NCs (QU and IN), because BACs and BP are lipophilic substances with lower polarity compared to NCs. In recovering NCs in the wash oil, the more difficult the components were to extract from the wash oil using the extraction solvent (an aqueous methanol solution), the easier it was to perform re-extraction. Therefore, contrary to the result of the above-mentioned equilibrium extraction operation of the wash oil, the m i,1 and (β i,IN ) 1 sequences of each component were 2MNA = 1MNA > BP > NA > QU > IN. Two kinds of NCs showed almost constant m i,1 and (β i,IN ) 1 regardless of E 0 /R 0 , but the m i,1 and (β i,IN ) 1 of other components, except NCs, showed a tendency to increase as E 0 /R 0 increased. In the range of E 0 /R 0 = 0.5-3, the m i,1 and (β i,IN ) 1 of other components, except NCs, showed a very high range of 1.3-3.7 and 15.9-47.6, respectively. From the fact that the m i,1 of IN showed the smallest value among six components, it was found that IN was concentrated in the raffinate phase by re-extraction using n-hexane. three kinds of BACs (NA, 2MNA, and 1MNA) and BP show significantly larger v those of two kinds of NCs (QU and IN), because BACs and BP are lipophilic s with lower polarity compared to NCs. In recovering NCs in the wash oil, the mo the components were to extract from the wash oil using the extraction solvent (a methanol solution), the easier it was to perform re-extraction. Therefore, cont result of the above-mentioned equilibrium extraction operation of the wash o and (βi,IN)1 sequences of each component were 2MNA = 1MNA > BP > NA > QU kinds of NCs showed almost constant mi,1 and (βi,IN)1 regardless of E0/R0, but th (βi,IN)1 of other components, except NCs, showed a tendency to increase as E0/R0 In the range of E0/R0 = 0.5-3, the mi,1 and (βi,IN)1 of other components, except NC a very high range of 1.3-3.7 and 15.9-47.6, respectively. From the fact that the showed the smallest value among six components, it was found that IN was co in the raffinate phase by re-extraction using n-hexane.   Figure 7a,b refers to a mixed extract phase which recovered through anol extraction used as the feed of re-extraction in this study. As predicted by NCs with strong polarity compared to the other four components are concentr raffinate phase, and it can be seen that the xi,n of NCs increases rapidly as n inc the other hand, the xi,n of the components with strong hydrophobic properties, e decreased rapidly as n increased. At n = 1-5, the xIN,n (i = IN) showed a very hig tration range of about 17.3-73.3%. Through this value, we were able to confirm of the re-extraction operation using n-hexane for the enrichment of IN. In Fig  found that the yi,n of NCs tended to increase sharply as re-extraction proceede remaining four components showed a nearly constant trend. It can be seen from that mi,n and (βi,IN)n tend to slightly increase as n increases in all components of except IN, which is the reference component of selectivity. In the range of n=1   Figure 7a,b refers to a mixed extract phase which recovered through the methanol extraction used as the feed of re-extraction in this study. As predicted by Figure 6, NCs with strong polarity compared to the other four components are concentrated in the raffinate phase, and it can be seen that the x i,n of NCs increases rapidly as n increases. On the other hand, the x i,n of the components with strong hydrophobic properties, except NCs, decreased rapidly as n increased. At n = 1-5, the x IN,n (i = IN) showed a very high concentration range of about 17.3-73.3%. Through this value, we were able to confirm the effect of the re-extraction operation using n-hexane for the enrichment of IN. In Figure 7b, we found that the y i,n of NCs tended to increase sharply as re-extraction proceeded, but the remaining four components showed a nearly constant trend. It can be seen from Figure 7c  Figure 2d,e show the gas chromatograms of the raffinate phase (R5) reco through a re-extraction run of n = 5, and a mixed extract phase (∑ ) , which is ture of each extract phase recovered through a re-extraction run of n = 1-5, respec Comparing the gas chromatogram of R5 in Figure 2d with that of the feed of the re-e tion operation of Figure 2c, the peak heights of two kinds of NCs (peak numbers 2 were much higher in the re-extraction run, but those of the other compounds were to be very small. In the gas chromatogram of the extract phase in Figure 2e, we cou that the peak height of each NC was slightly different from that of Figure 2c. The N extracted into n-hexane were very small compared to other components, so the height of each NC decreased, but those of other compounds did not show a sign difference. Through these gas chromatograms, it was possible to reconfirm the ef re-extraction using n-hexane on the concentration of IN presenting in a mixed e phase recovered through methanol extraction.  Figure 2d,e show the gas chromatograms of the raffinate phase (R 5 ) recovered through a re-extraction run of n = 5, and a mixed extract phase (∑ 5 n=1 E n ), which is a mixture of each extract phase recovered through a re-extraction run of n = 1-5, respectively. Comparing the gas chromatogram of R 5 in Figure 2d with that of the feed of the re-extraction operation of Figure 2c, the peak heights of two kinds of NCs (peak numbers 2 and 3) were much higher in the re-extraction run, but those of the other compounds were found to be very small. In the gas chromatogram of the extract phase in Figure 2e, we could see that the peak height of each NC was slightly different from that of Figure 2c. The NCs re-extracted into n-hexane were very small compared to other components, so the peak height of each NC decreased, but those of other compounds did not show a significant difference. Through these gas chromatograms, it was possible to reconfirm the effect of re-extraction using n-hexane on the concentration of IN presenting in a mixed extract phase recovered through methanol extraction. Figure 8 shows the composition (y IN ) and yield (Y IN ) changes of IN in solvent-free raffinate oil obtained at n = 5 for each operation run. The Y IN in each operation is based on the mass of IN contained in the wash oil, and this was calculated by Equation (2). As the extraction and the re-extraction operation progresses, y IN rapidly increased. The y IN in the wash oil used as the feed of this study was about 5.75%, but it was concentrated to about 73.3% as methanol extraction and n-hexane re-extraction proceeded. In the extraction operation, 100% of IN was recovered, but the loss of Y IN occurred due to the progress of the re-extraction operation, and about 79.1% of IN was recovered.

Changes of Composition and Yield of IN According as Each Operation
Molecules 2022, 27, x FOR PEER REVIEW extraction and the re-extraction operation progresses, yIN rapidly increased. The y wash oil used as the feed of this study was about 5.75%, but it was concentrated 73.3% as methanol extraction and n-hexane re-extraction proceeded. In the extrac eration, 100% of IN was recovered, but the loss of YIN occurred due to the progre re-extraction operation, and about 79.1% of IN was recovered. When comprehensively considering the results of this study, as described ab combination of methanol extraction and n-hexane re-extraction is expected to b the very useful combinations in recovering and concentrating IN from wash oil.

Recovery and Concentration Process of IN Containing in Wash Oil
The recovery and concentration process of IN from wash oil was investigate the experimental results obtained from the extraction and the re-extraction of thi As shown in Figure 9, the process suggested in this study was composed of one ex column, one re-extraction column, one washing column, and two distillation c Column 1 and 2, respectively, are an extraction column for recovering IN from t oil, and a re-extraction column for concentrating IN containing in the extract phas is upstream of column 1. Column 3 and 4, respectively, are a distillation column to concentrated IN oil from the raffinate phase, which is a downstream of column distillation column to recover the extract oil from the extract phase, which is an u of column 2. Column 5 is a washing column for removing methanol contained in finate phase, which is a downstream of column 1. When comprehensively considering the results of this study, as described above, the combination of methanol extraction and n-hexane re-extraction is expected to be one of the very useful combinations in recovering and concentrating IN from wash oil.

Recovery and Concentration Process of IN Containing in Wash Oil
The recovery and concentration process of IN from wash oil was investigated using the experimental results obtained from the extraction and the re-extraction of this study. As shown in Figure 9, the process suggested in this study was composed of one extraction column, one re-extraction column, one washing column, and two distillation columns. Column 1 and 2, respectively, are an extraction column for recovering IN from the wash oil, and a re-extraction column for concentrating IN containing in the extract phase, which is upstream of column 1. Column 3 and 4, respectively, are a distillation column to recover concentrated IN oil from the raffinate phase, which is a downstream of column 2, and a distillation column to recover the extract oil from the extract phase, which is an upstream of column 2. Column 5 is a washing column for removing methanol contained in the raffinate phase, which is a downstream of column 1.

Conclusions
To develop a simpler and more efficient novel process for the high-purity purifica tion of IN contained in the wash oil, this study, first of all, examined the concentration o IN present in the wash oil. For this examination, this study carried out an equilibrium methanol extraction to recover IN in the wash oil and an equilibrium n-hexane re-extrac tion to concentrate IN in the extract phase recovered through an equilibrium methano extraction. The combination of the methanol extraction and n-hexane re-extraction ap plied in this study showed very effective results for the concentration of IN in the wash oil, so it is expected to be one of the efficient combinations for determining the concentra tion of IN in wash oil. In the next paper, we will report the results of examining the high purity purification of IN through solute crystallization, using the 73.3% IN-containing raf finate phase recovered from the re-extraction operation of this study as the feed.

Conclusions
To develop a simpler and more efficient novel process for the high-purity purification of IN contained in the wash oil, this study, first of all, examined the concentration of IN present in the wash oil. For this examination, this study carried out an equilibrium methanol extraction to recover IN in the wash oil and an equilibrium n-hexane re-extraction to concentrate IN in the extract phase recovered through an equilibrium methanol extraction. The combination of the methanol extraction and n-hexane re-extraction applied in this study showed very effective results for the concentration of IN in the wash oil, so it is expected to be one of the efficient combinations for determining the concentration of IN in wash oil. In the next paper, we will report the results of examining the high-purity purification of IN through solute crystallization, using the 73.3% IN-containing raffinate phase recovered from the re-extraction operation of this study as the feed.