Catalytic Conversion of a ≥ 200 °C Fraction Separated from Low-Temperature Coal Tar into Light Aromatic Hydrocarbons.

A ≥ 200 °C fraction (CT200F) of low-temperature coal tar was prepared by a rotary film evaporator. The catalytic conversion experiments of CT200F and six model compounds were conducted on the pyrolysis gas chromatography-mass spectrometer. The yields of catalytic conversion products benzene, toluene, xylene, and naphthalene (BTXN) were analyzed by semi-quantitative analysis according to the chromatographic peak areas. Additionally, the possible formation pathways and mechanisms of the target products BTXN generated over different catalysts were investigated. The results show that the yield of aromatic hydrocarbons increases and the yield of acid compounds decreases during CT200F pyrolysis over ZSM-5, HY, USY, and β-zeolite compared with that of its non-catalytic pyrolysis, especially the yields of BTXN obtained over USY and β-zeolite increase by 128 and 108%, respectively. The pore structure of ZSM-5 is suitable to produce BTX, while the suitable acidity and pore structure of USY, HY, and β-zeolite are more beneficial for the selective preparation of naphthalene than that of ZSM-5. The conversion pathways of six model compounds into BTXN over zeolites were obtained, and the following conclusions can be drawn: The dehydroxylation effect of zeolites shows the order of ZSM-5 > HY > USY > β-zeolite. The catalytic effect of zeolites on the cracking and ring opening of PAHs in CT200F shows the order of β-zeolite > USY > HY > ZSM-5. The catalytic effect of catalysts on the cracking and aromatization of aliphatic compounds shows the order of ZSM-5 > β-zeolite > USY > HY. β-zeolite has an outstanding catalytic performance in the conversion of PAHs into naphthalene. ZSM-5 and HY can effectively remove phenolic hydroxyl groups in phenol and naphthol. During the catalytic conversion processes of the coal tar fraction and model compounds, the catalytic effect of the pore constructions of zeolites is more important than their acidities, which determines whether large molecules can enter and whether acid sites in non-micropores can be effectively utilized.

Introduction

Coal tar derived from coal gasification or coal pyrolysis contains abundant high value-added compounds, which are chemical raw materials.[1,2] China’s coal tar production is very considerable.[3] According to the data from the Ministry of Industry and Information Technology of the People’s Republic of China, it reached about 18.06 million tons in 2018. Faced with so much of the valuable coal tar resource, the method of processing and utilization of coal tar is rough, unsophisticated, and inefficient, especially for coal tar fractions with a high boiling point, which are not fully utilized.[4−6] The main components of coal tar high boiling fractions are aromatic hydrocarbons. They are very difficult to break down and mainly pyrolyzed to form coke and light oils/gases by the dehydrogenation and polycondensation reactions during the thermal cracking process, such as an industrial delayed coking process. However, the reaction selectivity of aromatic hydrocarbons is low.[7−9] If some high-efficiency catalysts can be used in the thermal cracking process of coal tar to increase the selectivity of major chemicals, coal tar deep processing will be promoted. However, according to the literature we have reviewed, the catalytic cracking of coal tar is rarely reported. Therefore, a deep research on the catalytic conversion technology of coal tar is urgently needed.

At present, many research studies are more focused on the catalytic conversion of gas-phase tar (volatiles) from coal pyrolysis.[10−12] It was found that metal oxides, carbon-based catalysts, and zeolite catalysts all had promoting effects on the directional conversion of coal tar as well as the yields of BTX.[13−15] Studies on coal pyrolysis showed that a large number of excessive research studies on the catalytic conversion of volatiles from coal pyrolysis (synchronous coal pyrolysis) have been investigated, and a variety of commercial and homemade catalysts have been discussed.[16−18] The performance of catalysts is difficult to evaluate correctly, and the structure–activity relationships of catalysts with the catalytic pyrolysis products are not accurate due to the complex coal pyrolysis process and uncertain pyrolysis products. In addition, the catalysis actually mainly affects pyrolysis volatiles. Therefore, our research team first proposes that the research on coal catalytic pyrolysis should be enhanced by the catalytic conversion of coal tar or its fractions. With this approach, the influence of coal pyrolysis process is weakened or removed, and this catalytic conversion process is more accurate and easier to be studied.[18,19] This approach provides a new research idea for the analysis of coal tar processing and utilization, as well as coal pyrolysis mechanisms.

Pyrolysis gas chromatography–mass spectrometry (Py-GC/MS) was widely used to analyze the thermal conversion processes of coal, biomass, oil shale, crude oil, and so forth.[20] It can reduce the possible systematic errors, for example, the mass loss of liquid products during condensation and collection and the secondary pyrolysis of primary products.[21] This analysis can reflect the compositions of pyrolysis products, and it is conducive to clarify the catalytic conversion mechanisms of samples, but it cannot reveal the real production rates of volatiles directly.[22] In order to solve this problem, the peak area results[23−25] were used to calculate the relative contents of volatiles. Even if an external calibration method was used to quantitatively calculate the real relative contents of major volatiles according to the peak areas, this method is not very accurate due to the lack of the actual amounts of the raw materials and gas–liquid–solid products. There is no doubt that Py-GC/MS remains an effective method to analyze the compositions of pyrolysis products, as well as to evaluate catalysts.

In our study, a ≥ 200 °C vacuum distillation fraction (CT200F) separated from low-temperature coal tar (LTCT) was prepared by a rotary film evaporator. ZSM-5, HY, USY, and β-zeolite were used as catalysts. The catalytic conversion experiments of CT200F and six model compounds (phenol, C22 alkane, naphthalene, naphthol, phenanthrene, and pyrene) were carried out by Py-GC/MS, and the yields and selectivities of the catalytic conversion products BTXN (benzene, toluene, xylene, and naphthalene) were analyzed by a semi-quantitative analysis based on the areas of chromatographic peaks. The formation pathways of the target products BTXN were investigated by the catalytic conversion of six model compounds over different catalysts, and these selected model compounds can effectively represent the different structural units of CT200F, and the possible formation mechanisms of BTXN from coal tar pyrolysis were further analyzed.

Experimental Section

Materials

Shaanxi Sihai Co, Ltd. in the Shaanxi Province of China provided the LTCT sample. The dehydration and deslagging process of the LTCT sample was pretreated according to the literature.[20] LTCT was separated by a rotary evaporator with a short path distillation unit system[26] (VDL70-4, VTA GmbH & Co. KG, Germany; evaporation surface area: 0.04 m2; condenser area: 0.1 m2) to prepare CT200F at 2 mbar. The basic properties and ultimate analysis of CT200F were analyzed and are summarized in Table . The typical compounds of phenol, C22 alkane, naphthalene, naphthol, phenanthrene, and pyrene were selected as model compounds for the catalytic conversion experiments. ZSM-5 zeolite (Si/Al = 46), HY zeolite (Si/Al = 5.4), USY zeolite (Si/Al = 5.4), and β-zeolite (Si/Al = 25) were provided by China Nankai University Catalyst Co., Ltd. to catalyze the catalytic conversion experiments of CT200F and model compounds.

Table 1

Basic Properties and Ultimate Analysis of CT200F

property		value
density, 20 °C/(g·mL–1)		1.05
viscosity, 80 °C/mPa·s		14.05
moisture, wt %		2.10
ash content, wt %		0.15
carbon residue, wt %		10.42
toluene insolubles, wt %		0.34
	Car	75.98
	Har	6.50
ultimate analysis, wt %	Oar(diff.)a	16.31
	Nar	0.90
	St,ar	0.31

Diff.: by difference, ar.: as received basis.

Characterization of Catalysts

N2 physisorption was used to analyze the structural characteristics of catalysts at about −196 °C (nitrogen saturation temperature) in an Autosorb-1 apparatus (TriStar II 3020M, Micromeritics, USA). Powder X-ray diffraction (XRD, D8 ADVANCE, Bruker, Germany) was employed to identify the crystalline phases of catalysts using Cu Kα radiation (λ = 0.15406 nm) at 40 kV and 150 mA. NH3–TPD (BELCAT II, MicrotracBEL Japan) was used to determine the acidities of catalysts. The testing conditions were referenced from the literature.[27]

Py-GC/MS Experiments

The chemical compositions of organic volatiles generated from the pyrolysis of CT200F and model compounds were investigated by a pyrolysis gas chromatography–mass spectrometer, which consists of a pyrolyzer (CDS, 5200, USA) and a gas chromatograph connected together with a mass spectrometer (GC/MS, Shimadzu QP2010 Plus, Japan), operated in an electron ionization (EI, 70 eV) mode.[28] The gas outlet of the pyrolyzer was directly connected to the injection port of the GC/MS. During this analysis, the sample (1 ± 0.01 mg) and the catalyst (2 mg) were laid out in layers in the quartz tube reactor in the following order: one layer of quartz cotton, one layer of catalyst, one layer of quartz cotton, one layer of sample, and one layer of quartz cotton.[18] Then, the sample was rapidly pyrolyzed at a heating rate of 20,000 °C.s–1 to 800 °C under helium (99.999%) and held for 10 s at this temperature, and the pyrolysis volatiles were brought into the GC/MS analyzer by a transfer line at 300 °C. Other operating conditions were set according to the literature.[20]

The chromatographic peaks of CT200F pyrolysis volatiles were identified by using the mass spectrum, with the MS library database of NIST08 and NIST08s. Moreover, a semi-quantitative analysis based on the peak area or peak area % of chromatographic peaks was conducted to represent the yields or selectivities of volatile products.[23−25]

Methods of Data Processing

The selectivity of a light aromatic (B, T, X, or N) and the carbon residue yields of spent catalysts were calculated by the following formulas.[29,30]where S represents the selectivity of light aromatics in BTXN; A1 represents the peak area of the one of benzene, toluene, xylene, and naphthalene obtained by GC–MS; ABTXN represents the total peak areas of benzene, toluene, xylene, and naphthalene; Y represents the yield of the carbon residue of the spent catalyst; m1 represents the mass of the carbon residue of the spent catalyst; and mt represents the mass of CT200F.

Results and Discussion

Characterizations of Catalysts

The structural properties of four catalysts are listed in Table . Based on Table , HY exhibits the highest BET surface area of 703 m2/g and the lowest pore diameter of 2.03 nm. ZSM-5 shows the lowest BET surface area and the lowest total pore volume. β-zeolite shows the highest pore volume of 0.44 cm3/g and the highest pore diameter of 3.53 nm.

Table 2

Characteristics of HY, USY, β-zeolite, and ZSM-5

catalyst	surface area, m2/g	total pore volume, cm>3/g	pore diameter, nm
HY	703	0.35	2.03
USY	700	0.35	2.04
β-zeolite	640	0.44	3.53
ZSM-5	350	0.20	2.21

XRD patterns of HY, USY, β-zeolite, and ZSM-5 are displayed in Figure . According to Figure , the characteristic diffraction peaks of ZSM-5 appear at 2θ = 7.96, 8.88, 23.07, 23.97, 24.45, and 25.91°, which match those reported for the typical MFI structure,[31] and the XRD pattern is very rich (The broad peaks are observed at 2θ = 22 and 25°.).[32] The intense diffraction peaks at 2θ = 7.7 and 22.5° belong to H−β zeolite.[33] The positions of the characteristic diffraction peaks of HY and USY are basically the same. They are located at 2θ = 6.20, 10.10, 11.91, 15.61, 18.65, 20.41, 23.69, 27.10, and 31.36°. The peaks of HY and USY are attributable to the FAU structure, which has the key features of strong reflections at 2θ = 6.39, 23.71, and 15.76°.[34] The XRD pattern of USY is a typical Y zeolite, which indicates the high framework crystallinity.[35] The high intensities of all peaks and low background lines indicate the high crystallinity of these catalysts.

Figure 1

XRD patterns of four catalysts.

The NH3–TPD profiles of HY, USY, β-zeolite, and ZSM-5 are shown in Figure , and the NH3–TPD fitting curves are listed in Figure S1 (see the Supporting Information). The peaks centered at around 211–236, 315–374, and 455–467 °C belong to weak acid sites, medium acid sites, and strong acid sites of NH3 desorption, respectively. All the catalysts tested exhibit three acid sites with different acid strengths. In addition, the peak temperature is usually associated with the strength of the acid sites. The position of the desorption peak corresponds to the acid center of the catalyst, and the area of the desorption peak corresponds to the surface acidity of the catalyst.[36,37]

Figure 2

NH3–TPD profiles of USY, HY, ZSM-5, and β-zeolite.

The results of acid distributions of USY, HY, ZSM-5, and β-zeolite are listed in Table . Combining Figure and Table , there are obvious NH3 desorption peaks of USY, HY, and ZSM-5 at a high temperature, while the intensity of the desorption peak of β-zeolite is relatively weak at a high temperature. The area of the high-temperature desorption peak of USY is the largest one, while the peak temperature of the high-temperature desorption peak of ZSM-5 is higher than that of the other three catalysts. The strengths of different acid sites have been listed in the order of HY > USY > ZSM-5 > β-zeolite for weak acid sites, β-zeolite > HY > USY > ZSM-5 for medium acid sites, and USY > HY > ZSM-5 for strong acid sites. HY has the highest content of weak acid sites among the four catalysts, whereas β-zeolite and ZSM-5 show lower contents of weak acid sites. The content of medium acid sites is the largest in β-zeolite and is the lowest in ZSM-5. USY has the largest content of strong acid sites.

Table 3

Acid Distributions of USY, HY, ZSM-5, and β-zeolite

		peak temperature, °C			amount of different acid sitesa
catalyst	total acid amounta	weak	medium	strong	weak	medium	strong
USY	1.00	215	315	455	0.38	0.26	0.36
HY	0.99	211	324	460	0.40	0.30	0.29
ZSM-5	0.54	236	366	467	0.28	0.01	0.25
β-zeolite	0.55	225	374		0.24	0.31

Values of the total acid amount and amounts of different acid sites are normalized integration values.

Product Distributions of CT200F Pyrolysis

The total ion chromatograms of volatiles generated from the non-catalytic and catalytic pyrolysis of CT200F are shown in Figure . The main components identified based on the peak area results are listed in Table S1 (see the Supporting Information). Based on the distribution characteristics of volatile products during CT200F non-catalytic and catalytic pyrolysis, the detected compounds were classified into six groups, containing aliphatic compounds, aromatic compounds, acid compounds, oxygen compounds, nitrogen compounds, and the others,[20] and the peak area % results of group compositions of volatile products are shown in Figure (Figure is obtained based on Tables S1.).

Figure 3

Total ion chromatograms of volatiles generated from non-catalytic and catalytic pyrolysis of CT200F.

Figure 4

Group compositions of volatile products generated from non-catalytic and catalytic pyrolysis of CT200F.

As shown in Table S1 and Figure , it can be found that the selectivity of aromatic compounds in CT200F pyrolysis products is about 57.48%, and they are mainly benzene, toluene, naphthalene, phenanthrene, and pyrene. The selectivity of aliphatic compounds is 24.59% in CT200F pyrolysis products, and C20–C24 alkanes have the selectivity of 30.07% in all detected aliphatic hydrocarbons. The selectivity of acid compounds in CT200F pyrolysis products is 11.90%. They mainly consist of phenol, methyl phenol, and naphthol, and their corresponding selectivities are about 1.34, 2.73, and 0.74%, respectively. In addition, a small number of oxygen compounds and other compounds (which are very poorly matched) are detected in CT200F pyrolysis products, whose selectivities are 2.69 and 2.04%, respectively.

In general, catalyst participation has a marked influence on the group compositions of volatiles generated from CT200F pyrolysis. The selectivities of aliphatic hydrocarbons gradually decrease by the catalysis of ZSM-5, HY, USY, and β-zeolite, and their peak areas are 34.48, 25.65, 22.28, and 13.29%, respectively. The selectivities of aromatic compounds gradually increase by the catalysis of ZSM-5, HY, USY, and β-zeolite, and their peak areas are 57.95, 66.97, 70.12, and 82.96%, respectively. The acid compounds are mainly concentrated in the pyrolysis products of CT200F non-catalytic pyrolysis, and only a small number of them are distributed in CT200F catalytic pyrolysis products over ZSM-5, HY, USY, and β-zeolite, and their peak areas are 4.59, 2.91, 4.93, and 0.12%, respectively.

In the catalytic pyrolysis products of CT200F over ZSM-5, the aliphatic hydrocarbons are mainly C20–C40 alkanes and also contain a certain number of olefins, such as butene and cyclopentadiene. It may be that some alkanes have been broken down.[20] The light aromatic hydrocarbons mainly include toluene, benzene, and xylene, and their selectivities are 11.22, 10.54, and 5.58%, respectively. In addition, there are many kinds of polycyclic aromatic hydrocarbons (PAHs), such as phenanthrene, fluoranthene, benzophenanthrene, and methyl perylene. Only fewer phenolic compounds have been detected, which may be due to the dehydroxylation of aromatic hydrocarbons.[38] During the pyrolysis of CT200F over HY, the aliphatic hydrocarbons are mainly C21–C36 alkanes and also contain a small number of olefins, such as butene and pentene, and their selectivities are 2.18 and 2.11%, respectively. The carbon numbers of aromatic hydrocarbons are mainly concentrated on C6–C10. The C10 aromatic hydrocarbons are mainly naphthalene, and its selectivity reaches 10.48%. The C6–C7 aromatic hydrocarbons are mainly benzene and toluene. In addition, their selectivities are 6.41 and 7.01%, respectively. The C8 aromatic hydrocarbons are mainly dimethylbenzene, and its selectivity is 4.07%.

The formations of olefins by the catalysis of USY and β-zeolite are similar to their formations over ZSM-5 and HY. The aromatic hydrocarbons derived from CT200F pyrolysis catalyzed by USY and β-zeolite consist of higher contents of benzene, naphthalene, and toluene, and their selectivities are 9.65, 10.45, and 10.69% (over USY) and 10.11, 10.04, and 9.54% (over β-zeolite), respectively. By comparing the distributions of pyrolysis products obtained before and after the addition of catalysts, it can be found that the addition of the four catalysts is beneficial to increase the selectivity of aromatic compounds and to reduce the selectivity of acid compounds. β-zeolite has a significant catalytic effect, which leads to the selectivity of aromatic hydrocarbons being increased by 25.48% and the selectivity of aliphatic hydrocarbons being decreased by 11.30%. The main reasons may be that aromatic compounds can generate hydrocarbon pools, as well as long straight-chain alkanes, by the reactions of thermal cracking, dealkylation, or ring opening. They can also generate the origin of catalytic carbons by the polycondensation reactions activated at the acid sites of catalyst pores directly.[18]

The selectivity of aliphatic hydrocarbons over ZSM-5 increases by 9.89% compared with that in the thermal cracking process without catalysts. This may be due to the formation of a large number of short-chain alkanes and simple olefins by the pyrolysis of many macromolecules over ZSM-5.[18,39,40] For example, the selectivity of butene obtained over ZSM-5 is 4.76%. The selectivity of acid compounds is 11.90%, of which phenolic compounds account for 74.52% of the total acid compounds during the non-catalytic pyrolysis of CT200F, followed by the catalysis of USY, ZSM-5, and HY. The selectivity of acid compounds over β-zeolite is lower. The phenolic compounds are rarely in the catalytic cracking products, and it can be concluded that the four catalysts have a good promoting effect on the removal of hydroxyl groups in phenolic compounds.[41]

Analysis of the Peak Area Results of BTXN

The method of data processing of the peak area results[23−25] was used to calculate the yields of light aromatic compounds to investigate the influence of the four catalysts on the formations of BTXN generated from the catalytic conversion of CT200F. Figure shows the yields of light aromatic compounds in volatiles generated from the catalytic conversion of CT200F based on the peak areas. According to Figure , the yield of BTXN increases significantly after adding the catalysts.

Figure 5

Yields of light aromatic compounds in volatiles generated from CT200F pyrolysis based on the peak area results.

In general, the acid sites of catalysts can provide abundant active centers for the catalytic reactions and can effectively promote the formations of light aromatic hydrocarbons from the cracking reactions and aromatization reactions of aliphatic hydrocarbons or the dehydrogenation of phenols.[42−44] Moreover, the catalytic effects on the yield of BTXN are as follows: USY > β-zeolite > ZSM-5 > HY > no catalyst. Compared with that in the pyrolysis without catalysts, the yields of BTXN obtained over USY and β-zeolite increase by 128 and 108%, respectively. According to the acidity results of Table , it can be observed that β-zeolite has a smaller total acid amount and no strong acid sites. The pore diameter of β-zeolite is 3.53 nm, which is larger than those of other catalysts. Therefore, the reason for the increase of BTXN obtained over β-zeolite may be that more macromolecules can pass through the pores into the β-zeolite catalyst and arrive at its acid sites, and then the corresponding catalytic cracking reactions take place. As for the good catalytic performance of USY, it may be attributed to the fact that there are enough strong acid sites on the surface of USY for catalytic conversion of CT200F pyrolysis volatiles, even though its pore size is a little smaller than that of β-zeolite.

Additionally, it can be seen that the yield of toluene is the highest in the volatile products of CT200F non-catalytic pyrolysis, followed by that of benzene, xylene, and naphthalene. Compared with that in the pyrolysis without catalysts, the yields of benzene and toluene obtained over ZSM-5 increase by 100 and 98%, respectively. The reason may be that more low molecular hydrocarbons are produced by the catalysis of ZSM-5, and they are aromatized to form aromatic compounds, such as benzene, toluene, and xylene.[13,14]

It is also worth noting that the catalysis of ZSM-5, HY, USY, and β-zeolite all can increase the yield of naphthalene. Compared to that in the pyrolysis without catalysts, the yields of naphthalene obtained over HY, USY, and β-zeolite all increase more than 300%, while the increase of naphthalene obtained over ZSM-5 is relatively small. The reason may be that ZSM-5 has the smallest amount of total acid sites among all catalysts, which might result from lacking enough active sites on its surface for the catalytic conversion of volatiles. On the other hand, ZSM-5 has abundant smaller pores (micropores), which is suitable to produce BTX by type selection. However, the formation of naphthalene may be in its mesoporous region.[45,46] Therefore, the suitable acidity and pore structure of USY, HY, and β-zeolite may be of more benefit for the selective preparation of naphthalene than that of ZSM-5.

Selectivity of Light Aromatic Compounds in BTXN

Through the study of Section , analysis of the peak area results of BTXN, the selectivities of BTXN over the four catalysts are investigated. The selectivities of these light aromatic compounds are calculated according to eq , as described in Section , Methods of Data Processing. As shown in Figure , the selectivities of benzene over ZSM-5, β-zeolite, USY, and HY are 33.25, 31.73, 26.68, and 22.92%, respectively.

Figure 6

Selectivities of light aromatic compounds BTXN over different catalysts.

It is indicated that ZSM-5 and β-zeolite are more conducive to the production of benzene, while HY has a poor catalytic effect. The selectivities of toluene and xylene obtained over ZSM-5 are 35.39 and 17.60%, respectively, which are higher than that obtained over the other catalysts. It can be explained by the fact that ZSM-5 has a better effect on the catalytic conversion of CT200F to form monocyclic aromatic hydrocarbons with C1 or C2 side chains. When HY and USY are used as catalysts, the selectivities of benzene, toluene, and xylene all show the same trends, while the selectivity of naphthalene has the opposite trend. In addition, the selectivities of xylene obtained over the four catalysts are much lower than that of the others in BTXN. Xylene has lower thermal stability than benzene and toluene. It can be concluded that HY, USY, and β-zeolite have good catalytic effects on the removal of polyalkyl side chains from aromatic rings.

Furthermore, the four catalysts show different catalytic effects on the generation of naphthalene. The selectivity of naphthalene over HY is the highest, reaching 37.47%. β-zeolite and USY also have higher selectivities for naphthalene, 30.70 and 29.55%, respectively, and ZSM-5 has the lowest selectivity for naphthalene, only 13.75%. As shown in Table , HY and USY have similar surface areas, total pore volumes, and pore diameters, while ZSM-5 has the smaller surface area and total pore volume. Therefore, the selectivities of naphthalene over HY and USY are not only related to the appropriate acidity and acid strength of the catalysts but also affected by their surface areas and pore diameters.

As shown in Table , some carbon deposition was generated in the four spent catalysts. The order of carbon residue yields of spent catalysts is as follows: β-zeolite > HY > USY > ZSM-5. Combined with Table , it can be found that a large amount of carbon is produced over β-zeolite, which may be due to its large total pore volume and diameter. These structural characteristics enable the macromolecules to enter the pore and react at the acid sites in the pore. While a small amount of carbon is produced over ZSM-5, this is mainly due to the small pore diameter of ZSM-5, which does not allow more large molecules to enter. Besides, the surface area of ZSM-5 is only 350 m2 g, which is the smallest one of all catalysts, resulting in the contacts between reactant molecules and acid sites being greatly reduced. Therefore, the catalytic cracking reactions of them can only take place on the catalyst surface. In addition, it also has to do with its low total acid amount and medium acid sites.[47] It can also be seen from the Py-GC/MS results of volatile products of CT200F that a large number of aromatic compounds are produced over β-zeolite and that a small number of aromatic compounds are generated over ZSM-5. The carbon residue yields of USY and HY are relatively close, and their yields of aromatic hydrocarbons in Py-GC/MS cracking products are also close, probably because the total amount of acid sites, surface areas, pore diameters, and total pore volumes of these two catalysts are similar.

Table 4

Carbon Residue Yields of Different Spent Catalysts

catalyst	ZSM-5	HY	USY	β
Y/wt %	13.68	26.96	23.78	31.27

Possible Formation Mechanisms of BTXN

In order to further explore the formation pathways of light aromatic hydrocarbons during the catalytic conversion process of CT200F, six types of compounds, which can effectively represent the different structural units of CT200F, were selected, including pyrene, phenanthrene, naphthalene, C22 alkane, naphthol, and phenol. Py-GC/MS experiments were carried out under the same conditions to focus on the formations of the target products BTXN over different catalysts. Figure shows the total ion chromatograms of volatiles generated from the catalytic conversions of model compounds. Figure shows the selectivities of light aromatic compounds in volatiles generated from the catalytic conversions of model compounds based on the peak area results. The main detected components are summarized in Tables S2–S5 (see the Supporting Information).

Figure 7

Total ion chromatograms of volatiles generated from the catalytic conversions of model compounds.

Figure 8

Selectivities of light aromatic compounds in volatiles generated from the catalytic conversions of model compounds, based on the peak area results.

According to Figure , naphthalene can produce benzene and toluene over the four catalysts. The highest selectivities of benzene obtained over ZSM-5, HY, USY, and β-zeolite are 6.02, 14.27, 12.48, and 16.05%, respectively. In addition, only ZSM-5 catalyzes the generation of xylene from the catalytic conversion of naphthalene, while β-zeolite, USY, and HY have no catalytic effect on the formation of xylene, which indicates that naphthalene may undergo dealkylation reaction to form benzene by catalytic cracking. The catalytic effect of β-zeolite on the formation of benzene is more obvious, and the selectivity of benzene is high, up to 16.05%.

No xylene has been generated during the pyrolysis of phenanthrene over all four catalysts. Only a small amount of toluene and benzene are produced over ZSM-5. HY and USY have the similar surface areas, total pore volumes, pore diameters, and acidity values. Their structures have FAU skeletons,[34] but they have greatly different catalytic performances in the catalytic pyrolysis of phenanthrene. There is no BTXN formation over HY, while benzene, toluene, and naphthalene are formed over USY. A possible reason is that USY has a partial mesoporous structure and HY does not have an obvious mesoporous structure. Phenanthrene mainly produces naphthalene over USY (the selectivity for naphthalene is 18.76%) and β-zeolite (the selectivity for naphthalene is 31.20%). Pyrene can produce benzene over all four catalysts. In addition to benzene and toluene, part of naphthalene is produced over HY. A large amount of naphthalene is generated by the catalysis of β-zeolite in the conversion processes of phenanthrene and pyrene, probably because the larger pore diameter (as shown in Table ) makes it easier for macromolecules to come in and react. Also, the micropores in β-zeolite skeletons are suitable for the production of N.[46]

C22 alkane can convert into BTXN by the catalysis of ZSM-5 and β-zeolite. No naphthalene has been generated over HY and USY. C22 alkane only produces benzene and toluene over HY. The most significant amounts of benzene have been generated from phenol pyrolysis over ZSM-5, HY, USY, and β-zeolite, and their selectivities are 18.39, 14.67, 9.36, and 7.34%, respectively. ZSM-5, HY, and β-zeolite all have catalytic effects for the formations of BTXN. USY catalyzes phenol to produce benzene and a small amount of naphthalene. Naphthol does not produce xylene by the catalysis of ZSM-5, HY, USY, and β-zeolite, but it produces a large amount of naphthalene. The selectivities of naphthalene obtained over ZSM-5, HY, USY, and β-zeolite are 27.12, 23.74, 19.00, and 19.01%, respectively. The results show that all the four catalysts have the ability to remove phenolic hydroxyl groups,[14,42] among which ZSM-5 and HY have the best removal effects.

By summarizing the above results, Figure presents the conversion patterns of model compounds into target products over different catalysts. As shown in Figure , the compounds in the blue box are the selected model compounds. In addition, naphthalene in the purple box is the model compound, which is also the target product of catalytic conversion of the other five model compounds. Taking C22 alkane and pyrene as examples, C22 alkane takes the pink line as its catalytic conversion route and the compounds pointed by the pink arrow as its target products, while pyrene takes the green line as its catalytic conversion route and the compounds pointed by the green arrow as its target products. Taking pyrene as an example, pyrene can produce benzene over the four catalysts, and the catalytic effects are as follows: USY > HY > β-zeolite > ZSM-5. The catalytic effects on the generation of toluene are as follows: HY > USY > ZSM-5. The catalytic effects on the generation of naphthalene are as follows: β-zeolite > HY. The four catalysts could not catalyze the conversion of pyrene into xylene. Possible formation mechanisms of BTXN generated from CT200F catalytic pyrolysis are shown in Figure .

Figure 9

Conversion patterns of model compounds.

Figure 10

Possible formation mechanisms of BTXN generated from CT200F pyrolysis.

Six conversion pathways are concluded in Figure ,[18,20,38] including (1) the dehydrogenation of alkanes, (2) the cracking of alkanes, (3) the dehydroxylation of phenolic compounds, (4) the cyclization of alkanes and alkenes, (5) the dealkylation of aromatic compounds, and (6) the cracking of aromatic compounds. According to Figure , the dehydroxylation effect of catalysts shows the order of ZSM-5 > HY > USY > β-zeolite, and the catalytic cracking effect of catalysts shows the order of β-zeolite > USY > HY > ZSM-5, and the catalytic effect for the aromatization of aliphatic compounds shows the order of ZSM-5 > β-zeolite > USY > HY. It can be seen from the conversion patterns of model compounds that acid compounds, such as phenol and naphthol, more easily remove phenolic hydroxyl groups to form benzene and naphthalene over ZSM-5 and HY.

Additionally, the removal efficiency of polycyclic compounds over β-zeolite is better than that of monocyclic compounds, probably because naphthol continues to crack after the removal of phenolic hydroxyl groups to produce more BTXN. The linear alkane is easily aromatized to form BTXN by the catalysis of ZSM-5 and β-zeolite. Naphthalene, phenanthrene, pyrene, and naphthol face difficulty in producing xylene over these catalysts. The catalytic activity of β-zeolite is very good for the pyrolysis of the five model compounds to produce naphthalene, which may be due to the large total pore volume, pore diameter, and medium acid sites of β-zeolite. These structures enable the macromolecules to enter into its pores and react at the acid sites in the pores, and then, more small molecules will be generated to enter into its micropores and to selectively convert into naphthalene. On the contrary, ZSM-5 can promote the production of benzene and toluene, but ZSM-5 has a poor catalytic effect in the catalytic conversion processes of pyrene and phenanthrene; the reason may be that the molecules of pyrene and phenanthrene are too big to get into the catalyst pores. Figure also shows that the selectivity of naphthalene obtained over ZSM-5 is very low, and the selectivities of BTX obtained over ZSM-5 are high. It may be that some compounds other than PAHs in CT200F are pyrolyzed to generate more BTX by the catalytic cracking.

In general, the acidity of zeolites is the main factor in controlling its catalytic activity,[48] which will affect the reaction process and the selectivity of the products.[47] USY has the highest total acid amount and a large proportion of strong acid sites, which results in a higher carbon residue yield. The yield of BTXN obtained over USY is the highest, which also proves that the acidity of USY is conducive to selectively produce BTXN. Based on the above research results, it is found that PAHs are the main compounds in coal tar and they are prone to coking by heat, which affects the acidity of zeolites, but this is not the determinant of coal tar catalytic conversion. The presence of β-zeolite is more favorable for the production of naphthalene, which may be due to its large total pore volume and pore diameter. These structures enable the macromolecules to enter into its pores and react at the acid sites in the pores. This is also the reason for the largest carbon residue yield of spent β-zeolite. ZSM-5 is more conducive to the formation of light aromatic hydrocarbons BTX. It may be that ZSM-5 has a small total pore volume (more small pores).[49,50] The micropore size is suitable for the formation of BTX. The catalytic reactions over ZSM-5 mainly take place in the interior of MFI skeleton pore, while PAHs (carbon) are formed in the pore interior of ZSM-5, except for its micropores. This is because the micropore size does not allow more large molecules to enter, and therefore, their catalytic cracking reactions are carried out on the surface of ZSM-5 or other large pores. Generally, the molecular size of PAHs in CT200F is larger than that of monocyclic aromatic hydrocarbons. The appropriate reduction of the pore diameter of zeolites can reduce the yield of PAHs (or the generation of carbon residue) and increase the yield of monocyclic aromatic hydrocarbons in catalyst pores.

Conclusions

The catalytic conversion experiments of CT200F and six model compounds over four catalysts at 800 °C were carried out by Py-GC/MS, and the possible formation pathways and mechanisms of BTXN were investigated. The yield of BTXN increases significantly after adding the catalysts. The catalytic effect on the yield of BTXN shows the following order: USY > β-zeolite > ZSM-5 > HY. The benzene selectivity of catalysts shows the order of ZSM-5 > β-zeolite > USY > HY, and the naphthalene selectivity of catalysts shows the order of HY > β-zeolite > USY > ZSM-5. Through the analysis of the conversion processes of six model compounds, it is found that the conversion of CT200F into BTXN involved six reactions, including the dehydrogenation of alkanes, the dehydroxylation of phenolic compounds, and so forth. In the catalytic conversion of coal tar, proper acidity of zeolites is necessary, but the most important thing is to take advantage of the acid sites of the pores. Suitable pores (larger than micropore) in zeolites are more favorable for macromolecule entry and catalytic conversion.