Bifunctional base catalyst for vacuum residue cracking gasification

Fuel Processing Technology 153 (2016) 1–8

Contents lists available at ScienceDirect

Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc

Research article

Bifunctional base catalyst for vacuum residue cracking gasification Ruiyuan Tang a, Yuanyu Tian a,b,⁎, Yingyun Qiao a,⁎⁎, Haifeng Zhou b, Guoming Zhao b a b

State Key Laboratory of Heavy Oil Processing, China University of Petroleum (East China), Qingdao, Shandong 266580, China Key Laboratory of Low Carbon Energy and Chemical Engineering, Shandong University of Science and Technology, Qingdao, Shandong 266590, China

a r t i c l e

i n f o

Article history: Received 18 May 2016 Received in revised form 11 July 2016 Accepted 23 July 2016 Available online xxxx Keywords: Bifunctional catalyst Cracking Light olefins Coke gasification Syngas

a b s t r a c t The residues cracking gasification was performed in a dual fluidized bed reactor by using calcium aluminate as the bifunctional catalyst, which was prepared by the solid phase method. Besides, carbon black assisted could contribute to improve the specific surface area of the catalyst. The performances of the reaction temperature and hydrothermal treatment catalysts on the products distribution for vacuum residue cracking were studied. It was found that solid base catalyst could promote to produce the products of light oils and light olefins, and _ co-producing the H2-rich syngas. The results showed that the C_ 2 + C3 olefinicity of higher than 58.0%, heavy oil conversion of above 92.0%, and carbon deposited of ca. 4.2 wt% were achieved by cracking over two selfmade solid base catalysts at 650 °C with a catalyst-to-oil ratio 7.0. Moreover, the catalyst with higher specific surface area displayed a better cracking performance than the one with lower specific surface area. Hydrothermal treatment base catalysts indeed decreased its activity, and thus led to higher liquid yield and lower gas yield. The carbon deposited on the catalyst was well gasified at 800 °C in steam-5.0 vol% oxygen. The H2 content and the H2/CO ratio were reached about 58.0 vol% and 4.5, respectively, with the CH4 content b 0.5 vol%. Besides, in comparison with an industry solid base catalyst, the coke gasification rate was obviously accelerated over two self-made solid base catalysts. Furthermore, the cracking performance of the solid base catalysts could reach basically stable via a few cracking gasification cycles. © 2016 Published by Elsevier B.V.

1. Introduction The proportion of unconventional oil resources reached about 14% of the world's oil supply. Extra-heavy oil and tar sand occupied about two thirds of unconventional oil resources. Therefore, processing these above-mentioned oils feedstock would have to produce a great amount of vacuum residue and atmospheric residue. At present, these residues are usually processed by the technologies of residue fluid catalytic cracking (RFCC) [1,2], coking [3–5], hydro-cracking [6] and visbreaking [7]. In some situations, several technologies could be integration used. RFCC technology is hard to directly processing the inferior feedstock due to the catalyst deactivation. Visbreaking technology is always used as a non-catalytic thermal pre-treatment process to reduce the viscosity of residual oils [8]. But it is difficult to treat high asphaltenes content of oil feedstock. Coking technology is widely used to treat the lowgrade residues due to wide feedstock adaptability and high operation reliability. However, it would have to produce plenty of the petroleum coke might containing high content of contaminants. In addition, hydrogen was the important material for pre-treatment the low-grade oil ⁎ Correspondence to: Y. Tian, State Key Laboratory of Heavy Oil Processing, China University of Petroleum (East China), Qingdao, Shandong 266580, China. ⁎⁎ Corresponding author. E-mail addresses: [email protected] (Y. Tian), [email protected] (Y. Qiao).

http://dx.doi.org/10.1016/j.fuproc.2016.07.022 0378-3820/© 2016 Published by Elsevier B.V.

feedstock and converting the cracking liquids of the residues into clean vehicle fuel and petrochemical material [9,10]. Now, in petroleum industry, hydrogen was mainly produced via catalytic reforming of naphtha or separation from the dry gas and LPG. The residues cracking gasification process would provide a novel route to convert the abovementioned residues into light products, and meanwhile co-producing hydrogen. And the truth, petroleum residue consists of large molecules and has high content of contaminants and low hydrogen content [11]. The desired products of cracking oil and gas, which are mainly the gasoline and diesel fraction, have high light olefins and H2-rich syngas. Thus, it has to contain two types of reactions. Namely, the residues are first converted into light products and also conducive to remove the contaminants via catalytic cracking. Then the H2-rich syngas and regenerated catalyst was produced via gasifying the cracking-generated coke. For integration of these two reactions, a dual fluidized bed reaction system was thus used and the catalyst particles could circulate between cracking and gasification reactions. Thus, the catalyst could provide not only the proper activity and stability for the residue cracking and coke gasification, but also the exothermic heat for the cracking reaction. The catalyst used in the cracking gasification cycle process was crucial. It should have not only proper activity for the residue cracking, but also high resistance to the contaminants. Both of these could ensure the stability of the catalysts, and produce the expected products at the

2

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8

possibly maximum degree. In regeneration, the used catalyst should have high the hydrothermal stability to the coke gasification in steamoxygen. There are many literatures studies on increasing light products yields and inhibiting the coke formation via optimizing or developing the residues cracking catalyst. The distribution of the cracking products was decided by the base strength of the catalyst. It was found that the addition of alkaline oxides could obtain the desired base strength and reactivity of the catalysts to reduce the coke and enhance the light products yields. Kolombos [12] found the yield of ethylene could reach 17– 23 wt% and no carbon deposited on the catalyst in 2 h by using MnO2 as the assistant. Basu [13] and Mukhopadhyay [14] found the reaction activation energy and the reaction temperature could efficiently reduce by using potassium impregnated calcium aluminate catalyst. Besides, solid base catalyst has been applied in the reactions of producing biodiesel [15–17] and coke gasification due to its good stability, high reactivity and production selectivity [18,19]. Wu [20] and Li [21] found the base catalysts were beneficial to convert petroleum coke into syngas and enhance the coke gasification rate. But it was rarely used solid base as bifunctional catalyst for the residues cracking gasification. The aim of this research is to study the performance of bifunctional solid base catalyst (calcium aluminate with different the base density) for both catalyzing the residues cracking and enhancing the coke gasification effects in a dual fluidized bed reactor. Three catalysts including an industry solid base catalyst and two self-made solid base catalysts with different base density were used to study the proper activity for vacuum residue conversion. An industry solid base catalyst with the lower base density was used as a reference to study the thermal cracking performance of vacuum residue. Silica sand is used as a reference to study the thermal cracking performance of vacuum residue. Two self-made solid base catalysts with higher base density were prepared on the basis of the industry solid base catalyst with the same crystal structure for vacuum residue conversion. The desired catalyst for the residues cracking and carbon deposited gasification has to maintain the catalyst properties steadily for the residues cracking. Hydrothermal treatment was performed for the fresh catalysts to adjust their cracking activity. Coke gasification process was conducted to remove carbon deposited on the catalysts, and the stability of the catalysts was also tested.

2. Experimental 2.1. Materials and reagents CaCO3 and Al2O3 were purchased from Tianjin Kemiou Chemical Reagent Co., Ltd, China, and calcined at 200 °C for 1 h. Hammett indicators and benzoic acid was purchased from Qingdao Jingke Chemical Reagent Co., Ltd, China. Carbon black (CB) purchased from Alfa Aesar and dried at 120 °C overnight, and used without further purification. Vacuum residue (VR) purchased from wintek refinery, China, and their properties were listed in Table 1. An industry solid base catalyst (calcium aluminate denoted as CA-0) was supplied by Zibo catalyst Factory, China, and their properties were exhibited in Table 2.

Table 1 Properties of wintek VR. Properties

VR

Density 20 °C/g·cm−3 Viscosity 100 °C/mm2·s−1 H/C ratio Carbon residue/wt% Elemental analysis/wt% C H S N O (by difference)

0.98 900 1.67 6.7 Group composition/wt% Saturates Aromatics Resins Asphaltenes

87.0 12.0 0.30 0.50 0.20

38.6 33.5 26.8 1.10

Table 2 Textural properties of solid base catalysts and Al2O3. Catalysts

Specific surface area/m2·g−1

Total pore Bulk Average pore volume/cm3·g−1 density/g·cm−3 diameter/Å

Al2O3 CA-0 CA-1 CA-1-800 CA-2 CA-2-800

112.5 3.2 9.6 8.3 25.9 23.6

0.25 0.01 0.02 0.02 0.05 0.04

0.78 1.50 1.45 1.40 1.30 1.31

95.4 92.8 91.6 105.5 70.9 95.3

2.2. Solid base catalyst preparation Bifunctional calcium aluminate catalyst was prepared by the solid phase method. The schematic diagram of the formation process of calcium aluminate catalyst with (a) and without (b) the addition of CB was shown in Fig. 1. CaCO3 and Al2O3 were firstly mixed with a CaCO3/Al2O3 molar ratio of 12:7, and they were evenly divided into two portions. In addition, 6.0 wt% CB was added in one of them. Then both of them were triturated respectively in the grinder for 3 min (3000 r·min−1) and calcinated at 1350 °C in argon for 2 h. And the carbon removal was performed at 600 °C for 2 h in air. These catalysts were denoted as CA-1 (without CB) and CA-2 (with CB), respectively. In addition, the fresh CA-1 and CA-2 catalysts were deactivated by hydrothermal treatment in steam to adjust its catalytic activity for VR cracking. The hydrothermal treatment temperature of 800 °C was adopted and maintained for 7 h. Thus, CA-1-800 and CA-2-800 was used to denote the hydrothermally treated CA-1 and CA-2 catalyst at 800 °C, respectively.

2.3. Cracking gasification experiment The residues cracking gasification was performed in a fluidized bed reactor. The schematic diagram of the test device was shown in Fig. 2. The test device consists of the oil feeding system, the temperature controlling system, the reaction system, as well as the separation and analysis system. The reactor was the core of the test device, which was made up of stainless steel with the total length of 800 mm and the inner diameter of 25 mm. The expanded section at top with the dimensions of L = 200 mm and D = 90 mm, and the reaction section of 60 mm in diameter. And a self-designed porous stainless steel distributor was used to fluidize the catalyst particles. VR and steam were preheated and was continuously into the reactor by a plunger oil pump. Of steam was used to atomize VR and purge the whole reaction systems, which was also part of the reagent gas for coke gasification. In addition, oxygen was used to form the reagent gas for the coke gasification and combusted the remaining coke. Nitrogen was used as the purging gas for each cycle process. For cracking, the fluidized bed reactor was first heated up to the needed reaction temperature under the atmosphere of nitrogen. Then VR and water was preheated to about 100 °C and 200 °C, respectively. Steam was first fed into reactor via an upward nozzle to fluidize the catalyst particles. When the temperature is stable, VR was fed into this reactor via another upward nozzle. And then vacuum residue and steam are mixed in the preheated section at the bottom of this reactor. Then the oil-steam mixtures pass through the upward nozzle at high speed and atomized into the tiny oil droplets, which was transported into the middle of the catalysts bed (such as silica sand and solid base catalysts) via a porous distributor, where are in turn vaporized and cracked through interacting with the catalyst particles. The cracking gas products were separated by an oil-gas separation system. The liquid products were cooled by a watercooled tube. The heavier oil was collected in the first collector, and then the lighter oil in the second collector was collected with the cooling water of 1 °C. The non-condensable gas contents were measured by the wet gas meter and analysis by GC.

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8

3

Fig. 1. Schematic diagram of the formation process of calcium aluminate catalysts with (a) and without (b) the addition of CB.

The spent catalyst was also regenerated in this fluidized bed reactor. Firstly, after turning off the feed of VR and steam, and nitrogen was switched to purge the whole reaction system for 30 min, and heated the reactor to the preset temperature of 800 °C (heating rate = 10 ° C·min− 1) in nitrogen. Subsequently, nitrogen was switched into the gasification reagents gas of steam-5.0 vol% oxygen. Of steam-5.0 vol% oxygen stands for steam plus 5.0 vol% oxygen introduced into the reaction system. The gas products could be measured by the wet gas meter and analysis by GC. When the VR cracking and coke gasification process

was not carried out, the coke content on the catalyst could be measured. And each cracking gasification cycle process was repeated at least twice, and material balance is over 95.0% with the relative error of the measurement b5.0%. 2.4. Characterization and analysis FTIR spectra of the samples were recorded by using Tensor 27 FTIR spectrometer in the range of 4000–500 cm−1. Each spectrum

Fig. 2. Schematic diagram of VR cracking gasification process (red line represented electric tracing band). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

4

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8

represented the average of thirty-two scans. The samples were grinded to powder and mixed with KBr (about 1.0 wt% to KBr). Textural properties were characterized by N2 adsorption measurements at − 196 °C with a Tristar II 3020 apparatus (Micromeritics Corp., USA). The test samples were degassed at 300 °C for 4 h under vacuum. Specific surface area was calculated by using the Brunauer–Emmett–Teller (BET) equation, and the total pore volume was calculated at P/P0 = 0.98. X-ray diffraction (XRD) patterns were recorded on an X-ray diffractometer X'Pert PRO MPD (PANalytical B.V. Netherlands). The analyzer was equipped with a CuKα radiator with k = 1.5408 Å working at 40 kV and 40 mA, and its scanning angle (2θ) was ranged from 5° to 75°at 8°/min. The base strength and total base number of the catalysts were measured by Hammett indicators method, including the bromthymol (H_ = 7.2), phenolphthalein (H_ = 9.8), 2,4-dinitroaniline (H_ = 15.0) and 4-nitroaniline (H_ = 18.4). Base strength of the used catalysts (about 0.3 g) was measured by the indicator-ethanol solution. The total base number of the catalysts was measured by the consumption of benzoic acid. Microscopic feature of the catalysts was observed by the cold field scanning electron microscopy (SEM) (S–4800, Hitachi, Japan) working with an accelerating voltage of 40 kV and 40 mA. Before the test, the tested catalysts were coated with gold to increase the electroconductivity, and the clear visibility of the catalyst surface morphology was obtained. The gas chromatography analysis was conducted to provide the detailed contents of the gas products. The non-condensable gas were divided into two groups, the dry gas (H2 and C1–C2) and the liquefied petroleum gas (C3–C5), and analyzed by using Lu Nan gas chromatography (GC-2060A, China) equipped with alumina plot packed column with the thermal conductivity and flame ionization detectors. Total gas yield was the sum of the dry gas and liquefied petroleum gas. The gas products mass yield was calculated according to total volume of gas products and the ideal gas law. The coke on the catalyst was burnt by pure oxygen and measured by a coke analyzer (HX -HW8B, Hua Xin, China). The distribution of the liquid products was tested with an Agilent 7890 A simulated distillation gas chromatography. The distillation fractions divided by boiling points, which including gasoline fraction in IBP-180 °C, diesel and VGO fraction in 180–350 °C and 350– 500 °C, respectively, and heavy oil fraction over 500 °C. Heavy oil fraction was defined as unconverted part in VR cracking. 3. Results and discussion 3.1. Catalyst characterization Table 2 listed the textural properties of all used solid base catalysts and Al2O3. The lower specific surface area and total pore volume of CA-0 (3.2 m2·g−1 and 0.01 cm3·g−1) and CA-1 (9.6 m2·g− 1 and 0.02 cm3·g−1) catalysts were detected. The reason might be attributed to the fact that the crystal agglomeration led to the pore structure sintering during calcination process. The specific surface area and total pore volume of CA-2 catalyst increased to 25.9 m2·g−1 and 0.05 cm3·g−1, respectively, while those of the hydrothermal treatment base catalysts slightly decreased to 8.3 m2·g−1and 23.6 m2·g−1, respectively. Average pore diameter of Al2O3 is comparable to that of CA-0, CA1 and CA-1-800 at around 90 Å with the typical mesoporous structure, while that of CA-2 catalyst was about 70 Å. Besides, the bulk density was 1.30 g·cm− 3 of CA-2 catalyst compared to that of above 1.45 g·cm−3 for CA-0 and CA-1 catalysts. It should be attributed to the fact that the textural properties of the catalysts varied a few, and caused the variation in the catalyst bulk density. Table 3 shows the base strength and the total base number of the catalyst detected by the Hammett indicators method [22,23]. Solid base catalyst exhibited different the base strength and total base number. The base strength of the hydrothermal treatment or the fresh CA1 and CA-2 catalysts was all ranged in 15.0 b H_ b 18.4, and total base number was reached about 4.0 mmol·g−1, while those of CA-0 were

Table 3 Base strength and total base number of solid base catalysts. Catalysts

Base strengtha/H_

Total base numbera/mmol·g−1

Base densityb/μmol·m−2

CA-0 CA-1 CA-1-800 CA-2 CA-2-800

7.2 b H_ b 9.8 15.0 b H_ b 18.4 15.0 b H_ b 18.4 15.0 b H_ b 18.4 15.0 b H_ b 18.4

0.05 4.0 3.2 4.1 3.4

15.6 416.7 385.6 158.3 144.1

a b

Measured by the Hammett indicators method. Base density = Total base number / Specific surface area.

7.2 b H_ b 9.8 and 0.1 mmol·g−1, respectively. Table 2 shows different the specific surface area of all used solid base catalysts, and thus led to different the base density of the catalysts (CA-0 = 15.6 μmol·m−2, CA-1 = 416.7 μmol·m−2, and CA-2 = 158.3 μmol·m−2). In addition, the base density of the hydrothermally treated catalysts was slightly decreased due to the decrease of total base number of the catalysts. This demonstrated that the self-made solid base catalyst might have stronger activity than that of industry solid base catalyst. Besides, the total base number of CA-2 catalyst has changed little compared to that of CA-1 catalyst. The reason might be due to the addition of CB had little influence on the total base number and only affect the textural properties of the catalysts. Fig. 3 shows the FTIR characterization results for all used solid base catalysts and Al2O3. FTIR spectra of all used solid base catalysts showed different status compared to that of Al2O3. A broad O\\H stretching vibration peak at 3460 cm−1 was observed corresponding to the O\\H bonds of free water and hydrogen bonds of the structural hydroxyl [24]. A weak O\\H bending vibration peak [25] at 1650 cm−1 was observed attributed to the absorbed H2O from air in sample or KBr. Besides, in comparison with Al2O3, two absorption peaks of all used solid base catalyst were observed at 570 and 1060 cm−1 corresponding to of Al\\O vibration [26]. A weak absorption peak observed at around 460 cm−1 due to the Ca\\O bond stretching vibration [27]. A broad absorption peak at 840 cm− 1 was observed, that was the Al\\O bond stretching vibration peak in the tetrahedron (AlO4) [28]. This demonstrated that calcium aluminate catalyst could form by the solid phase synthesis method at 1350 °C. Fig. 4 shows the XRD pattern of the fresh and hydrothermally treated solid base catalysts. It was found that all used solid base catalysts showed good crystallinity and diffraction peak intensity. For fresh base catalysts, Ca12Al14O33 crystal structure was observed. Besides, a small quantity of Ca3Al2O9 crystal structure was also found in the CA-1 and CA-2 pattern. The reason might be attributed to the Ca12Al14O33 crystal structure was prone to be generated than Ca3Al2O9 in this ratio

CA-2

CA-1

CA-0

1060 840

Al2O3

570 460

1650 3460

4000

3500

3000

2500 2000 1500 Wave number /cm-1

1000

Fig. 3. FTIR characterization of solid base catalysts and Al2O3.

500

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8

Hydrothermally treated base catalysts

Intensity (a.u.)

+ + +z + + + z + +

+ --Ca12Al14O33 Z--Ca3Al2O9

+ +z CA-2-800

+z

CA-1-800

Fresh base catalysts CA-2 CA-1 CA-0

10

20

30

40

50

60

70

2theta (degree) Fig. 4. XRD spectra of the fresh and hydrothermally treated solid base catalysts.

(n(CaCO3)/n(Al2O3) = 12:7). It was also found the basicity of Ca12Al14O33 was lower than that of Ca3Al2O9, which is consistent with the data exhibited in Table 3. The XRD pattern of the hydro -thermal treatment base catalyst (CA-1-800 and CA-2-800) has not changed much and only the peak intensity slightly strengthened. These results demonstrated that calcium aluminate catalysts exhibited good hydrothermal stability at 800 °C in steam. As mentioned above, Table 2 shows specific surface area of all used solid base catalysts, including CA-0 = 3.2 m2·g− 1, CA-1 = 9.6 m2·g−1, and CA-2 = 25.9 m2·g−1. CA-2 catalyst exhibited a relatively higher specific surface area than that of CA-0 and CA-1 catalysts. Different microscopic feature of all used catalysts could be observed

5

by SEM and was shown in Fig. 5. In Fig. 5(a), CA-0 catalyst consists of the irregular size structures particles, and some of them bonded together as aggregates. The SEM image of CA-1, CA-2 and CA-2-800 catalysts exhibited the similar particles morphology with CA-0 catalyst. These aggregates with the larger particles size were also observed in Fig. 5(b), (c) and (d). The particles size of the catalysts in the ranged of 1–10 μm, which was observed as the massive texture agglomerated from many small catalyst particles, ranged up to several tens of microns. It was also found the higher specific surface area of the used catalysts should obtain from the smaller size of these aggregates. Based on the SEM image of all used catalysts, it could be considered to be less-porous or even nonporous. Thus, the specific surface area of all used solid base catalysts should directly respond from the size of the catalyst particle. 3.2. Cracking over the fresh base catalysts The cracking test in this section shows all used solid base catalysts at different temperature with the VR feeding rate of 1.8 g·min−1, and maintained for 4 min. The cracking test in the reactor was varied in 30–60 g to change the catalyst-to-oil ratio. Thus, the catalyst-to-oil ratio of 7.0 was proper for VR catalytic cracking. VR feeding rate and superficial steam velocity are important in determining cracking performance and fluidization of the catalyst particles. The superficial steam velocity to ensure the full fluidization of the catalyst particles in this reactor was tested 10 times to obtain the optimal steam-to-oil ratio. Thus, a steam-to-oil ratio of 1.0 was selected to fluidize the catalyst particles and atomize VR. Products distribution of VR catalytic cracking over an industry solid base catalyst (CA-0) and two self-made solid base catalysts (CA-1 and + C_ olefinicity CA-2) was presented in Table 4. The C_ 2 3 _ _ (m(C2 + C3 ) / m(total hydrocarbon yield)), the heavy oil conversion

Fig. 5. SEM image of solid base catalysts: (a) CA-0 catalyst of 10 μm, (b) CA-1 catalyst of 5 μm, (c) CA-2 catalyst of 5 μm, (d) CA-2-800 catalyst of 5 μm.

6

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8

Table 4 Products distribution of VR cracking over three fresh base catalysts with a catalyst-to-oil ratio of 7.0. Catalysts

SDa

CA-0

Temperature/°C Gas yield/wt% _ C_ 2 + C3 /wt% _ C_ 2 + C3 olefinicityb/% Coke/wt% Conversion/% Liquid/wt% Gasoline/wt% Diesel/wt% VGO/wt% Heavy oil/wt%

600 10.6 3.2 30.6

600 15.7 5.9 37.5

650 18.8 8.0 42.8

700 22.6 11.0 48.5

600 23.0 12.9 56.1

650 26.9 14.9 57.8

700 30.5 17.4 58.1

600 24.6 13.7 55.8

650 29.7 16.4 59.3

700 33.2 19.8 59.7

5.8 64.6 83.6 8.7 11.6 27.9 35.4

5.2 73.5 79.1 12.2 15.0 25.4 26.5

5.2 79.8 76.0 15.4 17.5 22.9 20.2

5.3 84.2 72.1 19.7 18.1 18.5 15.8

4.4 89.0 72.6 25.9 21.4 14.3 11.0

4.1 92.3 69.0 30.3 21.5 9.5 7.7

4.2 94.5 65.3 33.5 20.3 6.0 5.5

4.6 90.2 70.8 27.1 21.3 12.6 9.8

4.3 94.0 66.0 32.0 20.6 7.4 6.0

4.4 96.6 62.4 35.3 19.7 4.0 3.4

a b

CA-1

CA-2

SD stand for silica sand. _ _ _ C_ 2 + C3 olefinicity = m(C2 + C3 ) / m(total hydrocarbon yield).

and the distribution of the cracking liquids were used to compare the catalytic cracking effect of silica sand and all used solid base catalysts. When silica sand is used, the liquid yield is 83.6 wt% and the light olefins yield is of 3.2 wt% at 600 °C and its conversion is around 65%. The mainly of the resulting cracking liquid is boiling points above 350 °C (VGO and heavy oil fractions). Furthermore, all the parameters (the gas yield, the _ C_ 2 + C3 olefinicity, the light oils yield and the heavy oil conversion) evidently increased over three solid base catalysts compared to those over silica sand. Moreover, the cracking effect over CA-1 and CA-2 catalysts was superior to CA-0 catalyst. Take the reaction temperature of 700 °C as an example. The yield of light olefins increased from 11.0 wt% over CA-0 to 17.4 wt% (CA-1) and 19.8 wt% (CA-2). The reason might be attributed to the increasing in the ethylene and propylene yields, which is associated with the cracking of butenes [29]. The increase in _ C_ 2 + C3 olefinicity with high total base number of CA-1 and CA-2 _ (i.e., C2 + C_ 3 olefinicity over CA-0 catalyst is 48.5% in comparison with 58.1% and 59.7% over CA-1 and CA-2 catalysts, respectively). The reason might be attributed to the changes in the density and strength of basicity of CA-1 and CA-2 catalysts (exhibited in Table 3). The conversion of heavy oil increased from 84.2% over CA-0 catalyst to 94.5–96.6% over CA-1 and CA-2 catalysts. Coke yield over CA-1 and CA-2 catalysts was slightly decreased compared to that over CA-0 catalyst. It was found that the coke was from dehydrogenation and condensation of asphaltenes and resin on the surface of the catalyst [30]. These results demonstrated that the self-made solid base catalysts could suppress the coke via interacting with coke and steam. The diesel fraction yield has not changed much. The gasoline fraction yield is increased from 19.7 wt% over CA-0 catalyst to 33.5–35.3 wt% over CA-1 and CA-2 catalysts associated with evidently change in the yields of VGO and heavy oil fraction. The reason should be attributed to about 15% of the heavy fraction of VR converted into the cracking gas and light oils products. Besides, it was also found that the cracking performance of CA-2 catalyst was superior to those of CA-1 and CA-0 catalysts. The gas yield over CA-1 and CA-2 catalysts exhibited an improvement with the cracking temperature increased from 600 to 700 °C. The reason should be attributed to VR cracking actually occurred both reactions of thermally and catalytically [31,32]. The thermal cracking reaction was prone to produce the small molecules, such as C1–C2 hydrocarbons by severe cleavage the molecules of VR. The catalytic cracking reaction (complied with the carbonium mechanism) was favor to generate more C3–C5 hydrocarbons than the C1–C2 hydrocarbons. Additionally, the reaction activation energy of the catalytic cracking reaction (124.3 kJ·mol − 1) was lower than the thermal cracking reaction (217.5 kJ·mol− 1) [14]. Therefore, the thermal cracking reaction was predominated at high temperature. And thus the gas yield was obviously increased with the reaction temperature increasing.

3.3. Cracking over the hydrothermal treatment base catalysts Table 5 shows the products distribution of VR cracking at 650 °C with different catalyst-to-oil ratio over the hydrothermally treated base catalyst. At a catalyst-to-oil ratio of 7.0 over CA-1-800 and CA-2800 catalysts (Test 1 and Test 2), the gas yield over CA-1-800 and CA_ 2-800 catalysts reached to 23.5–25.2 wt% and its C_ 2 + C3 olefinicity was 56.8% and 58.2%, respectively. In addition, the cracking liquid yield slightly increased to 70.6–72.4 wt%. In comparison with Table 4, these results reveal that the CA-1-800 and CA-2-800 catalyst showed lower basicity than CA-1 and CA-2 catalysts exhibited in Table 4, and thus provided much proper basicity and suitable for VR cracking. By improving the catalyst-to-oil ratio from 7.0 to 10.0 (Test 3 and Test 4), the liquid yield decreased further to 70.0 wt% (Test-3) over CA-1-800 and to 67.8 wt% over CA-2-800 (Test-4). Because of increasing catalytic activity of the catalysts, the heavy components conversion over CA-1-800 and CA-2-800 was a little bit lower than 95%, but still above about 92.0%. The gas yield was increased to 25.8–28.0 wt% from 23.5–25.2 wt%, and _ the C_ 2 + C3 olefinicity increased to 58.6–60.4% from about 57.0%. These results were similar with the catalyst-to-oil ratio of 7.0 over the fresh catalysts. This demonstrated that the cracking activity of the hydrothermally treated base catalyst indeed decreased. It is noteworthy that the coke yield was about 4.2 wt%, which was similar with those exhibited in Table 4. This demonstrated that solid base catalysts could efficiently slow down the catalyst deactivation via suppress the coke in cracking gasification cycle process, and it is an anticipated advantage for VR cracking and also for cracking gasification cycle process. And the cracking liquid was mainly gasoline and diesel fraction. Additionally, the yields of VGO and heavy oil fraction obviously increased compared to the fresh catalyst (exhibited in Table 4). As expected, higher catalyst-to-oil ratio would lead to higher gasoline fraction yield but lower VGO and heavy oil fraction yields. The reason might be attributed to the density of the excess of oxygen (such as peroxide ions or some similar entity) [33] active sites are evidently increased in higher catalyst-tooil ratio, which could strongly interact with hydrocarbon of vacuum residue molecule and weak the required energy of breaking C\\C bonds or C\\H bonds to form free radicals. And thus facilitate the effects of the residues cracking and dehydrogenation.

3.4. Coke gasification and catalysts stability The cracking-generated coke on the surface of catalyst could not only block the pores [34], but also further decrease the cracking performance due to decreasing total base number. Moreover, the heavy metals elements and heteroatoms might enrich into petroleum coke, and further lead to the catalyst poisoning deactivation. Thus, on-line regeneration catalyst by gasifying the carbon deposited, and it was great important to the cracking gasification cycle process. Carbon deposited on the catalyst was in time gasified by steam-5 vol% oxygen at 800 °C to produce syngas and regenerate catalyst.

Table 5 Products distribution of VR over the hydrothermal treatment base catalysts at 650 °C. Test No.

1

2

3

4

Catalysts Catalyst-to-oil ratio Gas yield/wt% _ C_ 2 + C3 olefinicity/% Coke/wt% Conversion/% Liquid/wt% Gasoline/wt% Diesel/wt% VGO/wt% Heavy oil/wt%

CA-1-800 7.0 23.5 56.8 4.1 90.8 72.4 28.4 23.5 11.3 9.2

CA-2-800 7.0 25.2 58.2 4.2 92.4 70.6 30.3 22.3 10.4 7.6

CA-1-800 10.0 25.8 58.6 4.2 92.1 70.0 30.2 22.9 9.0 7.9

CA-2-800 10.0 28.0 60.4 4.2 93.7 67.8 32.4 21.6 7.5 6.3

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8 Table 6 Gas composition from coke gasification over three solid base catalysts. Catalysts

CA-0 CA-1 CA-2 a

Gas composition/vol%

Coke conversion/%

H2

CO

CH4

CO2

C2–C3

54.2 58.1 59.3

14.5 12.9 12.6

0.3 0.5 0.4

30.1 27.7 26.8

0.9 0.8 0.9

7

Table 7 Products distribution of VR cracking over the regenerated catalysts. Ra/%·min−1

Catalysts

CA-1-800

CA-2-800

Original RC-1a RC-2a RC-3a Original RC-1 RC-2 RC-3 75.0 97.2 98.4

2.5 3.2 3.3

R(gasification rate) = m(coke conversion) / t(time).

Table 6 shows the gas composition and the coke conversion over three solid base catalysts. The gasification reaction was performed at 800 °C. The gasification time of 30 min was adopted and ended with no gas products detected. The composition of the generated syngas has not much difference for three solid base catalysts, suggesting the similar gasification performances over three solid base catalysts. Namely, a total H2 and CO2 content was reached to about 85.0 vol% (the H2 content reached about 59.0 vol%), and thus obtained the higher H2/CO ratio (about 4.5). And the CH4 content b0.5 vol%, which is similar to the results found by Wang [35]. The reason might be attributed to the methane-steam reforming reaction and the decomposition reaction was stronger than the methanation reaction at 800 °C. Besides, the coke conversion was obviously improved over CA-1 and CA-2 catalysts compared to that over CA-0 catalyst. Namely, it was achieved 75.0% coke conversion over CA-0 catalyst, whereas CA-1 and CA-2 catalyst could achieve about 98.0%. Thus, the gasification rate (R) of 2.5%·min− 1 was achieved on CA-0 catalyst, while this was about 3.2%·min−1 on CA-1 and CA-2 catalysts. The reason might be attributed to high base density and total base number of the self-made solid base catalyst could provide more oxygen active sites for facilitating the coke gasification. Furthermore, these excess of oxygen active sites (O2−) possess high activity for enhancing the coke on the catalyst surface could be quickly conversion. This demonstrated that CA-1 and CA-2 catalysts indeed enhanced its activity for gasification the coke, suggesting that the bifunctional characteristics of the self-made base catalysts (CA-1 and CA-2). Besides, oxygen was introduced into the carbon deposited gasification system, which can not only provide a source of oxygen for keeping catalytic activity of solid base catalysts, but also short the needed time for the catalyst regeneration. Table 6 shows the carbon deposited on the solid base catalyst was not realized a complete conversion. Thus, in this work, a two-step carbon deposited gasification process was used. Firstly, most of carbon deposited was gasified in the first step to produce the H2-rich syngas, and then the remaining carbon deposited on the catalyst surface was completely combusted by pure oxygen in the second step. The coke on the catalyst was gasified also represents the catalyst regeneration. In this section, VR cracking and the coke gasification were performed in a cycle process to test the stability of the catalyst. According to the cracking data exhibited in Tables 4 and 5, the cracking temperature and catalyst-to-oil ratio were selected to be 650 °C and 7.0 for CA-1-800 and CA-2-800 catalysts. The results of the residues cracking performances in 4 times of cracking gasification cycles over the original catalyst and the regenerated catalyst were exhibited in Table 7. Products distribution of VR cracking over the regenerated catalyst in a few cyclic processes was exhibited in Table 7. In comparison with the original base catalyst, the regenerated base catalyst cracking VR exhibited relatively the higher liquid yield but the lower heavy oil conversion. The heavy oil fraction yield in the cracking liquid of RC-1 was about 3.0 wt% points higher than that of original CA-1-800 and CA-2-800 catalyst. The gas yield and the heavy oil conversion slightly decreased about 1.7 wt% and 2.0%, respectively. The reason might be due to decreasing the basic catalytic sites in the gasification process, and thus reduced the catalytic activity of the catalyst. Thus, in comparison with the original catalyst, the regenerated catalyst shows the heavy oil fraction yield increased from 7.6–9.2 wt% to 9.7–12.5 wt%, the VGO fraction

Gas yield/wt% _ C_ 2 + C3 olefinicity/% Coke yield/wt% Conversion/% Liquid yield/wt% Gasoline/wt% Diesel/wt% VGO/wt% Heavy oil/wt%

23.5 56.8

22.0 55.7

22.3 56.1

21.8 56.4

25.2 58.2

23.5 56.9

23.1 57.2

22.9 56.8

4.1 90.8 72.4 28.4 23.5 11.3 9.2

4.1 87.5 73.9 23.5 21.3 16.6 12.5

4.2 88.2 73.5 23.4 21.4 16.9 11.8

4.0 87.9 74.2 24.0 21.8 16.3 12.1

4.2 92.4 70.6 30.3 22.3 10.4 7.6

4.3 90.3 72.2 26.4 21.2 14.9 9.7

4.2 89.8 72.7 26.0 21.1 15.3 10.2

4.1 89.5 73.0 25.9 21.4 15.2 10.5

a RC-1, RC-2 and RC-3 stands for the first, second and third times regenerated cycle process.

yield increased from 10.4–11.3 wt% to 14.9–16.9 wt%. And the gasoline fraction yield was correspondingly decreased to 23.4–26.4 wt% from 28.4–30.3 wt% via third times cycle process. Nevertheless, the diesel _ fraction yield, the C_ 2 + C 3 olefinicity and the coke yield have changed little. Based on all of the above results, it could conclude that the regenerated catalysts indeed decreased its activity via the regeneration process. And thus caused low heavy fraction conversion and gasoline fraction yield, high VGO and heavy oil fraction yield. But the cracking performances of the solid base catalysts could reach basically stable via a few cycle processes. These demonstrated that solid base catalysts could convert the residues into the light products and H2-rich syngas, and appeared to be suitable for using in VR cracking gasification.

4. Conclusions The residues cracking gasification over the bifunctional solid base catalysts could not only produce the products of light olefins and light oil via the residues cracking, but also co-producing the H2-rich syngas via gasifying the cracking-generated coke. In comparison with industry solid base catalyst (CA-0), the self-made solid base catalyst (CA-1 and CA-2) exhibited the higher residues catalytic activity. The results show _ that the C_ 2 + C3 olefinicity of above 58.0%, the carbon deposited of ca. 4.2 wt%, and the heavy oil conversion of N 92.0% at 650 °C with a catalyst-to-oil ratio of 7.0. Besides, the cracking performance of CA-2 catalyst was superior to that of CA-1 and CA-0 catalysts. The hydrothermally treated base catalysts indeed decreased its activity, and thus provided much proper activity for the residues cracking. The coke was well gasified in steam-oxygen at 800 °C with the H2 content was about 58.0 vol%, the CH4 content b0.5 vol%, and high H2/CO ratio (about 4.5). Besides, in comparison with the industry solid base catalyst, the gasification rate was significantly speeded up over two self-made solid base catalysts. The cracking performances of the regenerated catalysts indeed decreased a certain extent in the first cycle process, but it could reach basically stable via a few cycle processes. Solid base catalysts exhibited bifunctional characteristics of the residues cracking and the coke gasification, and this manifested the feasibility of its utilization for the residues cracking gasification.

Acknowledgements This study was conducted with National Natural Science Foundation of China (No. 21206185, 21576293 and 21576294), Key R&D Program in Shandong Province of China (No. 2015GGX107002), Qingdao Postdoctoral Applied Research Project (No. 2015202), and Scientific Research Foundation of China University of Petroleum for Recruited Talents (No. 2013010042).

8

R. Tang et al. / Fuel Processing Technology 153 (2016) 1–8

References [1] X. Dupain, M. Makkee, J.A. Moulijn, Optimal conditions in fluid catalytic cracking: a mechanistic approach, Appl. Catal. A 297 (2006) 198–219. [2] A. Guo, X. Zhang, Z. Wang, Simulated delayed coking characteristics of petroleum residues and fractions by thermogravimetry, Fuel Process. Technol. 89 (2008) 643–650. [3] H. Mizutani, Y. Korai, I. Mochida, Behavior of sulfur species present in atmospheric residue in fluid catalytic cracking, Fuel 86 (2007) 2898–2905. [4] S. Jasvinder, K. Surendra, O.G. Madhukar, Kinetic modelling of thermal cracking of petroleum residues: a critique, Fuel Process. Technol. 94 (2012) 13–144. [5] E. Furimsky, Characterization of cokes from fluid/flexi-coking of heavy feeds, Fuel Process. Technol. 67 (2000) 205–230. [6] S.K. Bej, Performance evaluation of hydroprocessing catalysts-a review of experimental techniques, Energy Fuel 16 (2002) 774–784. [7] J.B. Joshi, A.B. Pandit, K.L. Kataria, R.P. Kulkarni, A.N. Sawarkar, D. Tandon, Y. Ram, M.M. Kumar, Petroleum residue upgradation via Visbreaking: a review, Ind. Eng. Chem. Res. 47 (2008) 8960–8988. [8] Y. Zhang, D. Yu, W. Li, S. Gao, G. Xu, Bifunctional catalyst for petroleum residue cracking gasification, Fuel 117 (2014) 1196–1203 (Part B). [9] A.T. Jarullah, I.M. Mujtaba, A.S. Wood, Improving fuel quality by whole crude oil hydrotreating: a kinetic model for hydrodeasphaltenization in a trickle bed reactor, Appl. Energy 94 (2012) 182–191. [10] N. Hallale, F. Liu, Refinery hydrogen management for clean fuels production, Adv. Environ. Res. 6 (2001) 81–98. [11] Y. Zhang, D. Yu, W. Li, S. Gao, G. Xu, H. Zhou, J. Chen, Fundamental study of cracking gasification process for comprehensive utilization of vacuum residue, Appl. Energy 112 (2013) 1318–1325. [12] A. J. Kolombos, US, Pat. No. 4087350 (1978). [13] B. Basu, D. Kunzru, Catalytic pyrolysis of naphtha, Ind. Eng. Chem. Res. 31 (1992) 146–155. [14] R. Mukhopadhyay, D. Kunzru, Catalytic pyrolysis of naphtha on calcium aluminate catalysis -effect of potassium carbonate impregnation, Ind. Eng. Chem. Res. 32 (1993) 1914–1920. [15] B. Wang, S. Li, S. Tian, R. Feng, Y. Meng, A new solid base catalyst for the transesterifica -tion of rapeseed oil to biodiesel with methanol, Fuel 104 (2013) 698–703. [16] N. Viriya-empikul, P. Krasae, W. Nualpaeng, B. Yoosuk, K. Faungnawakij, Biodiesel production over Ca-based solid catalysts derived from industrial wastes, Fuel 92 (2012) 239–244. [17] W. Xie, H. Peng, L. Chen, Transesterification of soybean oil catalyzed by potassium loaded on alumina as a solid-base catalyst, Appl. Catal. A 300 (2006) 67–74. [18] Y.R. Xie, J. Xiao, L.H. Shen, J. Wang, J. Zhu, J.G. Mao, Effects of Ca-based catalysts on biomass gasification with steam in a circulating spout-fluid bed reactor, Energy Fuel 24 (2010) 3256–3261.

[19] C. Zhao, L. Lin, K. Pang, W. Xiang, X. Chen, Experimental study on catalytic steam gasification of natural coke in a fluidized bed, Fuel Process. Technol. 91 (2010) 805–809. [20] Y. Wu, J. Wang, S. Wu, S. Huang, J. Gao, Potassium-catalyzed steam gasification of petroleum coke for H2 production: reactivity, selectivity and gas release, Fuel Process. Technol. 92 (2011) 523–530. [21] Y. Li, H. Yang, J. Hu, X. Wang, H. Chen, Effect of catalysts on the reactivity and structure evolution of char in petroleum coke steam gasification, Fuel 117 (2014) 1174–1180 (Part B). [22] J.I. Take, N. Kikuchi, Y. Yoneda, Base-strength distribution studies of solid-base surfaces, J. Catal. 1 (1971) 164–170. [23] L.P. Hammett, A.J. Deyrup, A series of simple basic indicators. II. Some applications to solutions in formic acid, J. Am. Chem. Soc. 54 (1932) 4239–4247. [24] M.F. Zawrah, Investigation of lattice constant, sintering and properties of nano Mg– Al spinels, Mater. Sci. Eng. A 382 (2004) 362–370. [25] Y. Amenomiya, G. Pleizier, Alkali-promoted alumina catalysts: II. Water-gas shift reaction, J. Catal. 76 (1982) 345–353. [26] X.Q. Zhang, J.Z. Xiang, Y.M. Hu, A.M. Yang, J.H. Fang, L.T. Zhang, Study on preparation and IR absorbency of nanoscale Al2O3, China Ceram. 40 (2004) 24–27. [27] S.Z. Liao, B.W. Zhang, X.L. Shu, Y.F. Ouyang, H.W. Xie, Z.Y. Xu, The structure and infrared spectra of nanostructured MgO-Al2O3 solid solution powders prepared by the chemical method, J. Mater. Process. Technol. 89-90 (1999) 405–409. [28] V.M. Yanishevskii, Investigation of IR absorption spectra of binary calcium aluminate glasses and products of their crystallization, J. Appl. Spectrosc. 55 (1991) 1224–1228. [29] O. Awayssa, N. Al-Yassir, A. Aitani, S. Al-Khattaf, Modified HZSM-5 as FCC additive for enhancing light olefins yield from catalytic cracking of VGO, Appl. Catal. A 477 (2014) 172–183. [30] C. Xu, J. Gao, S. Zhao, S. Lin, Correlation between feedstock SARA components and FCC product yields, Fuel 84 (2005) 669–674. [31] K.N. Theologos, A.I. Lygeros, N.C. Markatos, Feedstock atomization effects on FCC riser reactors selectivity, Chem. Eng. Sci. 54 (1999) 5617–5625. [32] H.H. Gao, G. Wang, H. Wang, J.L. Chen, C.M. Xu, J.S. Gao, A conceptual catalytic cracking process to treat vacuum residue and vacuum gas oil in different reactors, Energy Fuel 26 (2012) 1870–1879. [33] A.A. Lemonidou, I.A. Vasalos, Preparation and evaluation of catalysts for the production of ethylene via steam cracking. Effect of operating conditions on the performance of 12CaO -7A12O3 catalyst, Appl. Catal. 54 (1989) 119–138. [34] S.J. Khang, J.F. Mosby, Catalyst deactivation due to deposition of reaction products in macropores during hydroprocessing of petroleum residuals, Ind. Eng. Chem. Process. Des. Dev. 25 (1986) 437–442. [35] J. Wang, Y.H. Yao, J.Q. Cao, M.Q. Jiang, Enhanced catalysis of K2CO3 for steam gasification of coal char by using Ca(OH)(2) in char preparation, Fuel 89 (2010) 310–317.